142 N. S. M. Mussafi & Z.

Ismail
DLSU Business & Economics Review 31(1) 2021, p. 142–157

RESEARCH REVIEW

A Review on Shariah Stock Portfolio Optimization

Noor Saif Muhammad Mussafi

UIN Sunan Kalijaga, Indonesia
Universiti Teknologi Malaysia

Zuhaimy Ismail
Universiti Teknologi Malaysia
zuhaimyutm@gmail.com

One of the essential concerns of investors of all time is to choose the best investment opportunities to capitalize on the value
of their investment in stocks. Its objective is to minimize risk and at the same time also maximize return. This refers to
portfolio optimization as the procedure of choosing the weights of a number of stocks to be included in a portfolio so that
the optimum objective is achieved. Several studies have been shown by economists, mathematicians, and practitioners that
provided sufficient and critical information for conducting stock portfolios. Approaches, methodologies, and techniques in
those researches have to be reviewed for purposes of synthesis of published research in this area, insight into how to carry
similar studies, and criticize gaps to be filled in the future research. In this report, we adopted a thematic review of stock
portfolio optimization theories and applications, including some current issues on Shariah stock portfolio. A comprehensive
overview of theories and approaches for stock portfolio selection, both fundamental and technical, is presented. The report
will end with an overview of the mathematical model (exact and heuristic) for solving stock portfolio optimization problems.

Keywords: Shariah stock portfolio, fundamental analysis, technical analysis, portfolio optimization, exact and heuristic.

JEL Classification: C44, C61, E61, G11, G32

Stock portfolio management provides important
information for government policy, authorities, or
investors to make an investment decision. There are
different approaches and methodologies for performing
such analysis. Recently, the optimization had prompted
research interest in developing the optimal financial
decision. Therefore, it is necessary to provide an
overview of the development in these studies.
This review highlights some contemporary issues on
stock investment and its characteristics in connection
with market fluctuation. The review also looked into
the current trend on Shariah stock investment, which
are claimed to experience a rapid evolution and
expansion of the Islamic financial services industry. It
has also received a wider acceptance and appreciation
beyond the conventional one.

Copyright © 2021 by De La Salle University

A Review on Shariah Stock Portfolio Optimization 143

We also adopted a thematic review of methodologies
and techniques used by several authors in forming a set
of stock portfolios based on fundamental and technical
analysis. The review considered the pros and cons of
two techniques from recent papers published in the
subject areas. The review style included analyzing the
underlying economic theory, mathematical theory and
methods, and its application.
Furthermore, this study attempts to review the
methodologies for solving stock portfolio optimization
problems. Two approaches that are frequently used are
heuristic and exact optimization models. A review of
these two approaches provides insight and guide in the
direction of future research interest. This review aims
at summarizing work on stock portfolio optimization.
Application of better techniques on forming a set of
stock portfolios and constructing the optimization
model are underway; therefore, this review only
provides a threshold in the understanding of stock
portfolio, stock portfolio selection, and stock portfolio
optimization.
Stock Investment
A share of stock, a type of financial investment
and literally referred to as “stock,” is a share in the
ownership of a corporation (Pentheny, 2009). Stocks
can be bought and sold at a price determined by the
financial success of the securities and the overall
demand for the stock securities. It is one of the most
versatile sectors in the financial system and plays
an important role in economic development (Yadav,
2017). An aggregation place of buyers and sellers
where stocks, bonds, or other securities are bought
and sold is known as the stock market.
The concept of a share market was first introduced
in France in the 13th century. The Dutch East India
Company was the earliest company to issue shares
on the Amsterdam Stock Exchange in 1602 (Osmani
& Abdullah, 2009). According to Siegel and Coxe
(2002), a century later, the first actively traded U.S.
stocks, floated in 1791, were issued by two banks: the
Bank of New York and the Bank of the United States.
Both offerings were enormously successful and were
quickly bid to a premium. However, they collapsed
the following year when William Duer attempted
to manipulate the market and precipitated a crash.
However, stock prices in the 1990s were dramatically
increased and influence stock markets across countries
(Duong & Siliverstovs, 2006). These situations
indicated that the stock market is commonly fluctuated
due to particular economic reasons.
Investment in stocks can be risky, particularly due
to stock price fluctuation. Fluctuation is said to occur if
there are a lot of shares for sale and no one is interested
in buying them, and the price will quickly fall (Vincent
& Bamiro, 2013). Fluctuations of stock prices and
stock indices result in a problem of uncertainty, which
is common to all stock markets. As presented in Table
1, there are several causes influencing the stock price
in the world as described by Yadav (2017), Kearney
and Daly (1998), Pentheny (2009), Vincent and Bamiro
(2013), and Siegel and Coxe (2002). Consequently,
many researchers suggested the importance of stock
portfolio risk management (Andersen, 2008; Ridha &
Alnaji, 2013).

Table 1. Summary of Analysis on Stock Price Fluctuation in the World

Country Causes Reference

Inflation rate, interest rate, financial leverage, corporate
India earnings, dividends yield policies, bonds prices, and Yadav, 2017
social and political variables
Australia Inflation rate, interest rates, and money supply Kearney & Daly, 1998
United Financial success of the company, the investor’s demand,
Pentheny, 2009
Kingdom and the confidence of investors
Monetary policy, a shift in mortgage lending toward the less
Vincent & Bamiro,
Nigeria creditworthy and marginal borrowers, and high operating
2013
costs in Nigeria
World conflict, gold price, monetary policies, and business
United States Siegel & Coxe, 2002
cycle
 during 2009-2013 the assets growth of Shariah stock markets, both Gulf Cooperation
144 N. S. M. Mussafi & Z. Ismail

ntries and others, grew, on average, by 17.5% or almost twice as the attainment of con
Shariah Stock Investment Global Shariah stock indices, represented by
Recently, the trend of investments based on ethical, 12 major active (as seen in Table 2), have gained
et (see Figure 1). Countries (both GCC and others) measured include Bahrain, Egypt, In
social, and environmental standards has expanded
substantially (Karim et al., 2014). Louche et al. (2012)
popularity due to the greater potential of growth and
profitability (Ho et al., 2014). Moreover, Hussain et
outlined responsible investment as a process through al. (2015) conveyed in their IMF working paper that
alaysia, Pakistan, Qatar, Saudi Arabia, Turkey, and United Arab Emirates.
which investors try to influence companies’ behavior
on a range of moral issues. Albaity and Rubi (2008)
during 2009-2013 the assets growth of Shariah stock
markets—both Gulf Cooperation Council (GCC)
concluded that investment in Shariah stocks is referred countries and others—grew, on average, by 17.5%
to the Islamic principles of transactions (Muamalat), or almost twice as the attainment of the conventional
Table 2: Global Shariah Stock Market from the
and therefore, in their view, Islamic investments also
US, European countries, and Asian countries
stock market (see Figure 1). Countries (both GCC and
fall into the type of ethical investment. Number
others) measured include Bahrain, of Indonesia,
Egypt,
Index Country of origin Year of origin
Kuwait, Malaysia, constituents Arabia,
Pakistan, Qatar, Saudi
Turkey, and United Arab Emirates.
w Jones Islamic Market Index (DJIMI) United States 2,374 1999
Table 2.
ndard & Poor Shari'ah Index (S&PSI)
Global Shariah Stock Market From the U.S., European United States
Countries, and Asian Countries 512 2007
ssell-Jadwa Shari'ah Global Index (RJSGI) United States 2,700
Number of Year of
2007
Index Country of origin
SE Islamic Index (FTSEII) United Kingdom constituents 2,700 origin 2009
yal Bank ofDow
Scotland Shari'ah
Jones Islamic Index
Market Index (RBSII)
(DJIMI) United Kingdom
United States 2,374 - 1999 2004
MI 150 Islamic Index
Standard (DMII)
& Poor Shari’ah Index (S&PSI) Switzerland
United States 512 150 2007 2008
Russell-Jadwa
ciete Generale Shari’ah Global
Wise Shari'ah Index Index (RJSGI)
(SGWSI) United States
France 2,700 600 2007 1999
FTSE Islamic Index (FTSEII) United Kingdom 2,700 2009
mbay Stocks Exchange Shari'ah 50 Index (BSESI) India 50 2008
Royal Bank of Scotland Shari’ah Index (RBSII) United Kingdom - 2004
karta IslamicDMI
Index (JII)
150 Islamic Index (DMII)
Indonesia
Switzerland 150
30 2008
2003
ala LumpurSociete
Shari'ah Index
Generale Wise (KLSI)
Shari’ah Index (SGWSI) Malaysia
France 600 30 1999 2000
ng Kong Islamic
Bombay Index (HKII)Shari’ah 50 Index (BSESI)
Stocks Exchange Hong Kong
India 50 16 2008 2007
Source: Ho, AbdIslamic
Jakarta Rahman,
Index Yusuf,
(JII) & Zamzamin (2014) Indonesia 30 2003
Kuala Lumpur Shari’ah Index (KLSI) Malaysia 30 2000
Figure 1: Growth of Conventional and Shariah Assets
Hong Kong Islamic Index (HKII) Hong Kong 16 2007

(in percent, compound average growth during 2009-2013)

Source: Ho (2014)

25

20

15
Conventional
10
Shariah
5

0
GCC Others Total

Source: (Hussain et al., 2015)

Source: (Hussain et al., 2015)
Figure 1. Growth of Conventional and Shariah Assets in Percentage (2009-2013)
me scholars also shown that performance of Shariah stock market tends to be better in

Malaysia Dow Jones Islamic Index (DJIM) and FSTE Bursa Malaysia Index (KLCI) (
A Review on Shariah Stock Portfolio Optimization 145

Some scholars have also shown that the performance
of the Shariah stock market tends to be better in terms
of return, for example, the Malaysia Dow Jones Islamic
Index (DJIM) and FSTE Bursa Malaysia Index (KLCI;
Karim et al., 2014; Reddy & Fu, 2014) and stable
against crisis (Abu Bakar & Ali, 2014; Reddy & Fu,
2014) compared the conventional one, for example,
Australian Stock Exchange (ASX).
Furthermore, many researchers also made a
comparison between the Shariah stock market and
the conventional stock market (Alam et al., 2017;
Geumei, 2018; Hussain et al., 2015; Osmani &
Abdullah, 2009). Table 3 exhibits their comparison
based on different views. However, in Reddy & Fu
(2014), later researchers have reported evidence of
a correlation between conventional market indices
between North American, European Union, Far
East, and Pacific markets compared to the Islamic
index returns (Dania & Malhotra, 2013). They
argued that the price of Islamic and conventional
securities corresponds to macroeconomic elements.
Consequently, both are almost the same except for
certain terms in trades mechanism and halal products
(Alam et al., 2017; Samra & Joseph, 2018).
As one of the rich countries in both renewable
(agricultural products) and non-renewable sources
(mining and minerals), Indonesia, even though still
behind Malaysia when based on Islamic finance
fundamentals (Yusof & Majid, 2009), has the
opportunity to develop the Shariah investment.
This prediction is based on the country’s narrow
fiscal deficits, low public indebtedness, healthy
economic growth prospects, the large scope of the
Indonesian economy, and the potential of the largest
Muslim population in the world. Lusyana and Sherif
(2017) added that the performance of the Indonesia
Shariah-compliant Stock Index (ISSI) and Jakarta
Islamic Index (JII) was promising by reason of the
increasing number of investors whose concern in the
ethical investment. Hence the Shariah investment
atmosphere in Indonesia might be interesting to
observe further.

Table 3. Comparison of Shariah and Conventional Stock Market based on Some Views

Basis of comparison Shariah Conventional

Existences It exists for Islamic corporations to raise capital
in a Halal way following the laws of the Shariah
Products Sukuk, stocks, sharia-compliant equities,
Islamic funds, equity funds Islamic exchange-
traded funds (IETFs), Islamic real estate investment
trusts (IREITs)
Places Traded in the same places as the conventional ones
since no Islamic only exchanges exist
Activities Only allowed to offer investment instruments for
projects that are useful and add value to the society
Rules Shariah principles: (1) free from all forms of Riba,
Al-Maysir, Al-Gharar, price controlling, Al-
Ihtikar, misinformation, and coercion, (2) halal
product (3) no pornography
Contract Mudharabah
Risk and Return Market risk sharing is essential.
So the return should be associated with the risk
It exists mainly to channel the wealth
of savers to those who can put it to
long-term productive use (to raise
capital)
Equities, derivatives (swaps, options,
futures, forward), and bonds
Available in approximately all the
countries
Offer investment instruments for
any profitable project whether it is
beneficent to the society or not; the
only important factor is how profitable
and how risky is it
It is not bound by any restrictions
Interest
Risk is better to be avoided.
Most derivatives are speculative and
developed to hedge against the risk
146 N. S. M. Mussafi & Z. Ismail

Stock Portfolio Selection ratios, has shown to be firm in discriminating the

performance of institutional investors (Mustilli et al.,
A portfolio as a set of stocks needs to be selected 2018). Overall, there is no such judgment of the two
appropriately to obtain the expected result. As the investment tools (fundamental or technical) that better
stock markets fluctuate dramatically, shares analysis select stock portfolios.
is one of the leading issues for investors to consider
the market trend and reduce gambling facets of stocks Fundamental Analysis
investment (Hooke, 2010). Basically, there are two Fundamental analysis is defined as a method of
known leading investment decisions tools that are evaluating a security by attempting to measure its
usually employed by investors in portfolio selection, intrinsic value. It deals with real data reports issued
that is, fundamental analysis and technical analysis. by corporate financial as evaluation by examining
Both of them can be used to determine the value of a related economic, financial, and other qualitative
stock and forecast its future performance (Cohen et al., and quantitative factors (Drakopoulou, 2016; Suresh,
2011; Wafi et al., 2015). 2013; Thomsett, 2005). More precisely, it attempts to
Many researchers compared the credibility between study everything that can affect the security’s value,
technical and fundamental analysis. According to including macroeconomic factors (like the overall
Moosa and Li (2011), Neely et al. (2010), and Wafi et economy and industry conditions) and specific factors
al. (2015), technical analysis outperforms fundamental (such as the financial condition and management
analysis. However, it contrasts with the study of Jakpar of companies). Campanella et al. (2015) added that
et al. (2019) that fundamental analysis is able to forecast fundamental analysis could predict the abnormal
and generate a positive return better than technical returnsAnalysis
Sequence of Fundamental to be triggered by dividend announcements
analysis in the food manufacturing in Bursa Malaysia. in the European securities market. Fundamental
Steps Activities
In addition, Silva et al. (2014) included a fundamental analysts also proclaimReference
Formula/Indicators
that financial results are the
analysis using financial1 :ratios
st
Economicinstead onlyrate,
of technical Inflation
Understanding dependable
interest rate, means for establishing the value of
Sathyanarayana
before dealing with the optimization algorithms. They a company. Nevertheless, Suresh (2013) argued that
analysis macroeconomic monetary policy, etc. & Gargesa,
claimed that financial ratios, which are quantitative one should capture not only the financial performance
measures, can intensely analyze the performance
environment and of a of the company but also 2018; the economic situation and
Suresh,
company in terms of profitability, liquidity, debt,
development and industrial environment. He
2013 labeled such examination
growth. Return on equity (ROE), one of the financial steps as EIC (economy, industry, company) framework.
(situation of a nation)

Table 4. Sequence of Fundamental

2nd: IndustryAnalysis
Analyzing the Industry life cycle, competitive Sabol et al.,

analysis prospects of the analysis of the industry, etc. 2013; Suresh,

Steps Activities Formula/Indicators Reference
industry to which the 2013
1 : Economic analysis Understanding macroeconomic
st
Inflation rate, interest rate, Sathyanarayana &
environment and development
company belong monetary policy, etc. Gargesa, 2018; Suresh,
(situation of a nation) 2013
(industry grouping)
2nd: Industry analysis Analyzing the prospects of the Industry life cycle, competitive Sabol et al., 2013;
industry to which
3rd: Company the company
Assessing the analysis of the
Debt, earning industry,
sale, etc.Jakpar et al.,
earning per Suresh, 2013
belong (industry grouping)
analysis financial performance share, equity, etc. 2019; Silva et
3rd: Company analysis Assessing the financial Debt, earning sale, earning per Jakpar et al., 2019; Silva
of a company al., 2014
performance of a company share, equity, etc. et al., 2014
th
4th: Ratio analysis 4 : Ratio the future
Predicting Predicting the future MohammedMohammed
et et al., 2020;
projections of earning net income Ghaeli, 2017; Silva et
analysis projections of earning ROE = al., 2020;
totalequity al., 2014
Ghaeli, 2017;
share priceof stock
P/E = Silva et al.,
earning per share
2014
Notes: ROE = Return on equity and P/E = Price earnings ratio
Notes: ROE = Return on equity and P/E = Price earnings ratio
A Review on Shariah Stock Portfolio Optimization
Moreover, Mohammed et al. (2020) enriched
an extra step so-called ratio analysis. There are
many ratios that can be used, but they are all
categorized into one of three groups: liquidity
ratios (determine how quickly the company will
fulfill its obligations at maturity without incurring
a loss), profitability ratios (measure and evaluate
the ability of a company to generate income (profit)
relative to revenue), and leverage ratios (indicate
the utilization of borrowed money or level of debt).
Table 4 illustrates the sequence of fundamental
analysis along with supporting information desired.
Although there are different and varying concepts
related to the fundamental analysis, the economists
agree on the same objective, that is, to forecast future
earnings and the true value of the assets so that
investors can make investment decisions at the right
time to gain profit. In addition, from all formulas of
ratio analysis displayed in Table 4, the price/earnings
ratio (P/E) is perhaps the most primary measurement of
a stock’s value. As a fundamental indicator, according
to Ghaeli (2017), it is normally implemented as a
metric to compare individual stocks and the market as a
whole relative to historical valuations. Accordingly, the
portfolio selection might be executed by considering
the properties on fundamental analysis as base choices,
but it seems that it needs to be verified by other
analyses.
Technical Analysis
Unlike fundamental analysis that focuses on
intrinsic value, technical analysis deals with the
behavior of the market, which only focuses on the
price action, such as the historical price of a certain
period. The aim of technical analysis is to help
investors select the proper portfolio and the best time
to buy and sell the stock by spotting demand and
supply levels plotted on a chart (see Eiamkanitchat et
al., 2017; Jakpar et al., 2019). The term “best” refers
to the most appropriate period to either perform (buy
or sell) or not perform (hold) the trading by employing
one of some common useful methods in technical
analysis such as exponential moving average (EMA),
simple moving average (SMA), relative strength index
(RSI), the stochastic oscillator (STO), the moving
average convergence divergence (MACD), and the
average directional index (ADX). Eiamkanitchat
et al. (2017) recommended EMA; the trend of the
prices indicates by the exponent values those plotted
147
on the price chart, as a suitable method in resulting
the best average earning compared to MACD, STO,
and RSI. In addition, de Souza et al. (2018) revealed
that stock markets in BRICS member nations that are
technically analyzed based on SMA, EMA, and its
combination are profitable in terms of return, whereas
Jakpar et al. (2019) preferred to employ MACD to
test its correlation with stock returns.
The portfolio selection, firstly introduced by
Markowitz (1952), is well-known as the main problem
in the finance literature and investment practice;
because the future returns of shares are not known at
the day of the investment decision, the problem is one
of decision-making under risk measure (Geambasu et
al., 2013; Kulali, 2016). The most important aspect of
Markowitz’s model was his description of the impact
on portfolio diversification by the number of securities
within a portfolio and their covariance relationships
(Mangram, 2013). Substantially, it was argued that an
infinite number of “efficient” portfolios exist along a
curve defined by three variables: standard deviation,
correlation coefficient, and return. The efficient-frontier
curve consists of portfolios with the maximum return
for a given level of risk or the minimum risk for a given
level of return (Todoni, 2015).
Such remarkable invention is based on modern
portfolio theory (MPT), which focuses on risk
measures by employing variance or standard deviation.
Unfortunately, a considered risk has been a debatable
topic. Many scholars such as Rockafellar and Uryasev
(2016), Rom and Ferguson (2009), Roman and Mitra
(2009), and Mangram (2013) suggested new risk
measures even though nowadays MPT remains the
most commonly used measure in portfolio selection
practices. Furthermore, with the intention of risk-
adjusted return on MPT, Sharpe (1994) then proposed
the Sharpe ratio (Kolbadi & Ahmadinia, 2011; Pilotte
& Sterbenz, 2006). It is the ratio of the excess return
to the standard deviation of that return. Additionally,
Jack Treynor (1965) provided an alternative reward-
to-risk ratio, that is, the Treynor ratio. It is the ratio of
the excess return to the systematic risk of that return
and reliant upon a portfolio’s beta (market who decides
risk; Pilotte & Sterbenz, 2006).
Another theory to fill the gap of MPT, particularly
on the market reality, is termed post modern portfolio
theory (PMPT). Geambasu et al. (2013) claimed
that it includes the behavior of the investor in the
computation of risk measure, considering the risk
 TR =
βp 2006; Todoni, 2015

Return Table 5 . Differences Between

Symmetrical MPT and PMPT Based
Asymmetrical Mangram,on Some 2013; Perspectives
148 Table
Table 5 .5 Differences
. Differences Between
Between MPT MPT and and PMPT PMPT Based
Based onon Some Some Perspectives
N. Perspectives
S. M. Mussafi & Z. Ismail
distribution Perspectives (penalizing upside and MPT(penalizing downside PMPT Rom & Ferguson, Reference
Perspectives
Perspectives MPT MPT PMPT PMPT Reference
Reference
Risk measure
downside Standard
part) deviation
Table 5. Differences Between MPT and PMPT Based on Some Perspectives part only) Downside risk 2009; Roman Roman & & Mitra,
Risk
Risk measure
measure Standard Standard deviation
deviation Downside
Downside risk risk RomanRoman && Mitra,
Mitra,
Mitra, 2009 2009; Todoni, 2015
Perspectives MPT PMPT Reference
Risk measure Standard deviation Table 6 Downside risk
2009;2009; Todoni,
Todoni,
Roman & Mitra, 2009;
2015 2015
Risk-adjusted
Benchmark rate Risk-free rate, driven Sharpe ratio (SR) Sortino ratio
Minimum acceptance Rasiah, 2012;Todoni, 2015 (SoR) Geambasu et al.,
Risk-adjusted
Risk-adjusted
Risk-adjusted return ResumeSharpe Sharpe
Sharpe
Between ratio ratio
VaRratio
(SR) (SR)(SR)
and cVaR Stand Sortino
Sortino
SortinoonratioSomeratio
ratio (SoR)
(SoR) Geambasu
Views
(SoR) Geambasu
Geambasu et
et et
al.,al.,
al., 2013;
return by government SR =
rp − rf rate (MAR), SoR driven
=
rp − rby f Todoni, 20152013; Kolbadi &
Kolbadi & Ahmadinia, 2011;
rσ rσ
Views return return VaR SR SR = =
p − prprf− rf
cVaR
SoRSoR = =
p −rprf− rf
d
2013;2013;
Reference
PilotteKolbadi
Kolbadi
& & &2006;
Sterbenz,
σpσp market σ d σ d Ahmadinia,
Todoni, 2015 2011;
Formula VaRα ( X ) = min{z | FTreynor Treynor ratio
ratio (TR)
* (TR) ∞ Ahmadinia,
RockafellarAhmadinia, & 2011; 2011;
Assumption Not considering
X ( z) ≥ α}
Treynor
Treynor ratio
cVaR α ( X ) =
ratioConsidering
(TR)(TR) −∞ ∫ z dFXα ( z )* , where ,
investor Geambasu et al., Pilotte & Sterbenz,
TR =
rp − rf Pilotte
Uryasev, Pilotte &&
2016; Sterbenz,
Sterbenz,
used investor expectation rpβ−prprf− rf expectation 0, when z < VaR α ( X ) 2013; 2006;
Mangram, Todoni, 2015
TRTR = = 
Return distribution Symmetrical FX ( z ) =  FXupside
α
β p β p(penalizing ( z ) − α Asymmetrical (penalizing Roman 2006;2006;
&Todoni,
Mangram, Todoni,
Mitra,2013; 2015 2015&
Rom
 1 − α , otherwise.
Return Symmetrical
and downside part)  Asymmetrical
downside part only) 2013 Mangram,
Ferguson,2013;2009; Roman &
Return
Return Symmetrical
Symmetrical Asymmetrical 2009;
Asymmetrical Mitra,
Mangram,
Mangram, 2009
Sarykalin et
2013;
2013;
Tested by distribution
Benchmark rateResulting in a lower
(penalizing
Risk-free rate,upside Resulting
driven byand in aMinimum
higher acceptance
(penalizing Cheng, 2001
downside rateRom & Ferguson,
Rasiah, 2012; Todoni, 2015
distribution government
distribution (penalizing
(penalizing upside
upside andand (penalizing (MAR), driven by market
(penalizing downside
downside al., Rom2008
Rom && Ferguson,
Ferguson,
bootstrapping
Assumption used return downside
Not considering return
part)investor partConsidering
only) investor 2009; Roman
Geambasu et &
al., 2013;
expectation expectation Mangram, 2013
Function Non-convex anddownside downside part)part) and continuous
Convex part part only)only) 2009;
AlHalaseh 2009; Roman
etRoman
al., & &
simulation
Tested by bootstrapping Resulting in a lower return Resulting in a higher Mitra,Cheng, 2001
2009
simulation return
profile discontinuous 2016; Mitra,
Mitra, 2009
2009
Rockafellar
Notes:
Notes: prBenchmark
= portfolio return, rate
r f = Risk-free
risk free rate, σ p rate,
= driven
standard Table p6
deviation, Minimum
Table
β = 6
portfolio acceptance
beta, σ d = downside Rasiah,
risk 2012;
Benchmark
Benchmark rate rateRisk-free
Risk-free rate,
rate,driven
driven Minimum Minimum acceptance
acceptance & Uryasev,
Rasiah,
Rasiah, 2016)
2012;
2012;
by government
Resume Between VaR
Resume Between VaR and cVaR Stand on Some Views and ratecVaR (MAR), Stand driven
on Some by Todoni,
Views 2015
Average Lower Between VaRbyand
Table 6. Resume by
government
government
cVaR Higher
Stand on Some Views rate rate(MAR),(MAR), driven
driven by byTodoni,
RockafellarTodoni, 2015
& 2015
Views Views VaRVaR 14 market cVaR cVaR Reference
Reference
of losses
Views VaR market
market cVaR Uryasev, 2016;Reference
Formula
FormulaAssumption
Formula )Not considering* Considering ∞
investor Geambasu et al.,
Rockafellar
Rockafellar
Rockafellar &
&& Uryasev,
z | FXz(|zF) X≥(αz }) *≥ α }cVaRcVaR
∫ αz dF*α ( z )* , where ,
∞
VaRαVaR ( X )α=( Xmin{ = min{
Assumption
Assumption Not Not considering
considering
α ( X )α (
= X
∫ ) =z dF
Considering
X ( z ) X, where ,
−∞ Considering
−∞
investor
investor Sarykalin
Geambasu2016;
Geambasu Roman
et al., al.,&
et et al.,Mitra,
used investor expectation expectation 2013;
Uryasev,Mangram,
2009; Sarykalin
Uryasev,2016; 2016; et al., 2008
used
used investor
investor expectation
expectation 0,
0, expectation when
when z < VaR α (2008
z < VaR ( X )
 expectation
 X )α 2013;2013; Mangram,
Mangram,
α 
) X=(z )F=X (z )F−X (αz ) − α
FXα ( zF
, otherwise.
2013 Roman
Roman & Mitra, & Mitra,
 1− α 1 − ,α otherwise.
Losses VaR does not control cVaR accounts  for losses 2013
Sarykalin 2013et al.,
Function
Tested by Resulting
Non-convex and discontinuous
in a lower Resulting
Convex and continuous
in a higher Cheng,
2009; 2001
2009; Sarykalin
Sarykalin
AlHalaseh et
et al.,et2016;
scenario scenarios
profile Tested
Tested bybyexceeding VaR in
Resulting
Resulting exceeding
in a lowerVaRResulting
a lower Resulting in ina higher
a higher 2008 Cheng,
Cheng, 2001
2001& Uryasev,
Rockafellar
bootstrapping return return al.,
al., 2008
2016) 2008
Averagebootstrapping
Deviation bootstrapping
VaR
of is not a deviation
Lower return
return cVaR is Higher strongreturn return
competitor to Roman Rockafellar
& Mitra, & Uryasev,
losses simulation
Function Non-convex
Function Non-convex and and Convex
Convex and continuous and continuous AlHalaseh
2016;
AlHalaseh Sarykalin
et al.,et etal.,
al., 2008
Losses simulation
measure simulation
measureVaR does not control scenarios standardcVaR deviation accounts for losses exceeding 2009; Sarykalin
Sarykalin etetal., 2008
scenario
profile Notes:
profile discontinuous
exceeding
discontinuous
r p = VaR
portfolio return, r f = risk free rate, σ VaR
p = standard deviation, β p = portfolio beta, σ d2016;
= 2016;
downside Rockafellar
Rockafellar
risk
Deviation Notes: VaR is
Notes: rp not =aportfolio
deviation
=rpportfolio return, rmeasure
return, f =r frisk
= risk
freefree σcVaR =isstandard
p σ= pstandard
rate,rate, strong competitor
deviation,
deviation, =to
β p β= pportfolio al.,beta,
standard
portfolio
beta, σ2008
dσ
Roman
= ddownside &
riskMitra,
= downside risk 2009;
measure deviation & Uryasev,
Sarykalin
& Uryasev, 2016) et 2016)
al., 2008
Notes:
Notes: * X = random variable with the cumulative distribution function, FX ( z ) = P{ X ≤ z}, and confident level α ∈ [0,1]
Average
Average Lower Lower Higher Higher14 Rockafellar
Rockafellar & &
14 14
of losses
of losses Uryasev,
Uryasev, 2016;
2016;
Portfolio Optimization
Sarykalin
Sarykalin et al.,
et al.,
A Review on Shariah Stock Portfolio Optimization 149

as to the chance that the investment return is less be combined directly or indirectly in considerations
than the minimum return expected by the investor of the expected value of the portfolio’s rate of return
from his portfolio. Compared with MPT, Lee and as well as the return’s dispersion and possibly other
Eid (2018), Rom and Ferguson (2009), and Roman measures of financial risk.
and Mitra (2009) argued downside risk metrics Portfolio optimization is the technique of finding the
of PMPT as the essential difference in favor of best portfolio for the investors, given the available set
more appropriate risk measures for asymmetric of portfolios and the investor’s tolerance for risk. It is
distributions of returns. Moreover, to calculate often called mean-variance optimization (MVO). The
risk-adjusted return based on downside risk, Sortino term refers to the expected return of the investment,
proposed a modified version of Sharpe ratio, named and variance is the measure of the risk associated
Sortino ratio (Rasiah, 2012). It includes minimum with the portfolio. The mean-variance concept is
acceptance rate (MAR) and downside risk as constructed initially by Markowitz (1952). It deals
variables to be considered (Rom & Ferguson, 2009). with a point of view of portfolio investment decision
Table 5 reviews in more detail the differences of that allows investors to plan, predict, monitor, and
methods between MPT and PMPT. decide on the desired risk and return. The fundamental
The next popular risk measure of portfolio goal of MVO is to optimally allocate investments
performance is value at risk (VaR). It considers between different assets. Kulali (2016) asserted that
the confidence level of maximum losses. Although investors who are risk-averse and efficient portfolios
VaR is one of the most accepted measures of must meet two conditions, namely minimizing the
risk, according to AlHalaseh et al. (2016), it has variance of portfolio for a given expected return or
undesirable mathematical characteristics such as a maximizing the expected return for a given variance.from the conv
is derived
is derived from the convex quadratic programming (Fu, 2019; M
lack of sub additively, convexity, and coherent isisderived only
derivedfrom The
fromthe first
theconvex condition
convexquadratic results
quadraticprogramming in the
programming(Fu, problem (Fu,finding
of 2019;
2019;Mussafi, Mussafi,201 201
when it is based on the standard deviation of normal a minimum variance portfolio of the stocks that
is yields
toderived
consider anainvestor withquadratic
ofa fund to isinvest to consider in n different an investor stoc
distributions. To overcome this lack, the other isisto consideris
risk
toconsider ananinvestor at least
investor from with
with the
target convex
aafund
fundvalueto
toinvest
invest the in in nprogramming
expectedn different
differentreturn. stocks.
stocks. (Fu,Let 2019;
Let riri b
is derived
measure
is derived
introduced from the
from the
by convexconvex quadratic
Rockafellar quadratic and Uryasev programming
programming (Fu, 2019;
Mathematically,
(Fu, 2019; Mussafi, Mussafi,
the model 2012).
2012). of MVO The basic
The basic
is derived idea here
idea here
from
associated with the rate of return for associated
stocks withi =the 1,...,rat
(2016), i.e., Conditional Value at Risk (cVaR) was
associated
associated
thetoconvex
is
with
with consider
the
the rate
quadratic
rate an
of
of
programming
investor
return
return with
for
for a (Fu,
stocksfund2019;
stocks ito
,
i ,for
for
Mussafi,
invest
i i == inin,nn,for
1,...,
1,..., , different
and
and defin
defin
,
nst
performed. is derived
It is thean from
average the convex
of with quadratic
someaapercentage programming of in(Fu, 2019; Mussafi, 2012). The basic idea here
is to
is to consider
consider an investor
investor with fund to
fund to invest
invest in 2012). different
nn different The basic stocks.
stocks. ideaLet hererri is
Let betothe consider
random
i be the random variable
an variable
investor
( rni,) for
T
the worst-case loss scenarios and approximately with
=TT( r1stocks.
z associated a fund toT invest in n different stocks. z
rn )with Set
thethe µi rate (ri of ), mreturn ( µTT1 , µfor Let= rrµi1 be the
, r)2T,..., .=Set =µi∑n
, r2 ,...,Let 2 ,..., stocks , and cov( i z=)1,...,
is to consider an investor with a fund to invest in n different ri .be= random E=variable
equal,associated
at the samewith
associated with
confidence the rate
the ratelevel, to
of return
of return
zz =
the average =
for
for
((11 stocks
r r, ,
r r ,...,
,...,
stocks
22 rnn)
r )ii,, for
random. .=Set
= Set µµ
for ii =variable
i1,...,
=1,...,
i E E
= =( (
r r ),), mm
associated
and define
nn,, i iand
((µµ , µ
, µ
11 with
define
,...,
,...,
22 the µµ )
the rate) ,
random
the nnrandom
,and
and cov(
cov(
of return n
z
vector
vector
z)) =
for= ∑ ∑ . . If
If w w == ((ww1
of losses greater associated thanwith or equal
the ratetoofVaR return(Roman
for stocks&i, for i =stocks 1,..., n, and i fordefine i =T 1,...,n,
the random and define
vector the random vector
Mitra, 2009; Sarykalin et al., 2008). Furthermore, z = ( r1 , r2 ,..., rn ) . = Set µi E= (ri ), m ( µ1weights , µ2 ,..., µnin )T an,nand cov( z )nTh
portfolio. =
zz == ((rr11,,rr22and
Rockafellar ,..., rn2)),...,..r=
T
= ( r1rr,Uryasev
z,...,
n
T
n=Set
Set
)
T
.= µ
Setµ
(2016),
ii
µ i E
E=
E=(
= (r
rr
i
ii
),
), m
m
Sarykalin
), m ( µ1(
(weights
,µweights
µ
µ
11
2 ,, µ
µet
,..., µ
22
,...,
,...,
n al.
)T
, in
in
µ
µ
and
nn
a
)
weights
a
) portfolio.
,
cov(portfolio.
, and
and
and
in
cov(
cov(
a
zThe
z )
portfolio.
The
) == ∑
1rate
∑rate
. 2 If
TT z ) = ∑ . If w = ( w , w ,..., w )T is a set of
. If
If ofww
ofn
The
== (
return
return
( w
rate
w 11 ,, w
of
of
w
of
22
return
,...,
this
this
,..., ww nn
)TT
portfolio
portfolio
)
of
is
is
this
a
a set
set
r
set r
portfolio
∑
∑
= = of
of
of r r ww
ii ii
r=
is
is ∑also
also
i =1
a
(2008), and AlHalaseh et al. (2016) showed that CVaR weights in a portfolio. The rate of return of this portfolio i i
= 1
=1 n
portfolioweights
r = ∑ riTwi in aalsoportfolio. Thevariable
T rate
with of with return of this T portfolio r =
∑
n
is superior to weights VaR ininoptimization a portfolio. Theapplications.
rate of return of this Table isisalso a random
nna random variable mean
mean m w and variance
TT mean i =m wrand variance w ∑random w. If µvariable b is the targeted value of
weights in
weights
6 summarized inthe
aa portfolio.
portfolio.
method’s The The rate rate of
characteristics of return
return
mean
mean
between ofmm
of this
thiswwand portfolio
portfolio
and
andvariance
variance
variance
1
∑
r == wwTTrri∑ wiw
iw
∑ iwis
is also
If µµbabaisis
also
.. If random
the
the targeted
the variable
targeted
targeted value
value
value with
with ofof of expected
expected ri =r
VaR and cVaR. expected return, ii==11 then an optimal portfolio is a portfolio
mean mT w and variance wT ∑ w. If µb is the targeted value
portfolio
of expected
is
return, then an optimal
T a portfolio that solves the portfolioquadratic
following is a portfolio progra th
portfolio
portfolio is
is aamean
that
portfolio
portfolio m w
solves the
that and
that variance
following
solves
solves the
the wT ∑ w
quadratic
following
following programming µb is the(QP).
. Ifquadratic
quadratic targeted value
programming
programming (QP
(Q o
mean
mean m
Portfolio Optimization
m TT
w
w
portfolio and
and is variance
variance
a portfolio ww
that TT
∑∑
solves w
w ..
the If
If µ µ
following is
is the
the targeted
targeted
quadratic value
value

programming of
of
(QP). expected
expected return,
return, then
then an
an optimal
optimal
bb
1 T 1
portfolio
11 TT min is a portfolio w ∑ w that solves the following quadratic min wT ∑prog w
1 T
min
min w w ∑∑ w w 2
x 2
There
portfolio are
portfolio is always
is aa portfolio
min
portfolioxtwo 2
w sides
∑w to
that solves
that an
solves the investment,
the following
followingxquadratic quadratic
x 22
x
programming (QP).
programming (QP).
(1)
namely risk and return. s.t. mTAs w ≥ µab , egeneral
w = 1, andrule w ≥ 0, in the s.t. Tm T
w ≥T µb , eT w = 1, and w ≥ 0, (1) s.t. mT w ≥ µb ,
T 1
T

return must expect more ss.t.t. . mm ww≥≥µµbb,min

TT
economy, one who seeks ,ee ww==w 1,1, and
∑andw ww≥≥0,0,
1 1 more T x 2
min e denotes w who
T
∑w wvector
(1) e denotes the vect
risk, and conversely, where min one w the
∑ huntsof for ones.less risk will where
xx 22 where
where e edenotes
denotess.t. of the
m theT vector of
w ≥ µb , eof
vector T ones.
= 1, and w ≥ 0,(1)
wones.
gain less return. TheHaving objective of such a problemwhere
where is e
toedenotes
denotes thethe vector
vector of ones.
ones.
ss..ttat.. themmTTw wa≥≥portfolio
µµbtime
b,, e eTTw wconstructed
= 1, and properly
1, and ww ≥≥ 0, 0, either by fundamental Having aorportfolio technical analysis,
constructed one properly either by
minimize risk and, same to=maximize return. Having a portfol
portfolioHaving fundamental
where Having
e denotes oratotechnical
portfolio
the constructed
theanalysis,
vector of ones. one properlycan see the either by funda
This refers tocan see the emphasis
portfolio optimization on howas to the
solveprocedure Having
optimization, aa portfolio
portfolio
that is, how constructed
constructed
allocate properly
properly
assets either
either byby fundamental
fundamental or or t
where ethe
where
of choosing edenotes
denotes
weights the
the of vector
vector
a number of ones.
of ones.
of stocks to be emphasis on how to solve portfolio optimization, that
optimally. In recent years, mathematical programming can see
howthe
is,techniques to emphasis
have
allocate become on
a the
vitalhow
tools to to solve can see the
portfolio emphasis ont
optimization,
included in a portfolio so that the optimum can objective
can see see the the emphasis
emphasis Having
onon how how totoassets
solveoptimally.
portfolio
solve portfolioInoptimization,
constructed
portfolio recent
properly
optimization, years, either
that
that is,by fund
is, how
how
for certain condition Havingisaaobtained.
Having
support assets portfolioThe
portfolio
allocation, constructed
constructed
making condition would
process properly either
properly
and being mathematical
either by
gradually by fundamental
applied in programming
fundamental or
financial decision. techniques
or technical
technical analysis,
It is have
analysis, one become one
optimally.
optimally. In recent years, mathematical In recent
programming yea
techni
optimally.
optimally.
one of the Operations Research techniques Incan
In
which seek see the
recent
recent
to maximize emphasis
years,
years,
or on howofprogramming
mathematical
mathematical
minimize a function to solve portfolio
programming optimization
techniques
techniques have
have
can see
can see the
the emphasis
emphasis on on how
how to to solve
solve portfolio
portfolio optimization,
optimization, that that is,
is, how
how to to allocate
allocatesupport
the assets
the assets
assetsgradually
allocationa
many variables subject to a set of constraints enforcedsupport assets
optimally. Inallocation,
recent making
years, process and
mathematical being
programming
support
support assets by the nature
assets allocation, of the problem
allocation, making being
making processstudied
process andand being
being gradually appliedtech
gradually applied in
in f
optimally.
optimally. In
In recent
recent years,
years, mathematical
mathematical programming
programming techniques
techniques have
have become
become vital
vital tools
tools to
to
one of which
the Operations
and integrally restrictions on some or all of the variables. In contrast to other mathematical
one of the Operations Research techniques seek to maR
150 N. S. M. Mussafi & Z. Ismail

Yousfat (2015) stated that by having a set of portfolio

vital tools to support assets allocation, making process authors compared the results from the proposed model
and being gradually applied in financial decision. It is with the results from mean-variance model and MAD
one of the Operations Research techniques which
the fund allocation by employing the quadratic programming
model. According to their results, the proposed model
seek to maximize or minimize a function of many provides better returns than the mean-variance and
variables subject to a set of constraints enforced by main MAD objective
models. of QP is to reduce the variance, which consi
the nature of the problem being studied and integrally Yousfat (2015) stated that by having a set of
restrictions on some or all of the variables. In The Yousfat
QP model
portfolio,(2015)
onestated
is that
defined
could by having
minimize a set
as follows.
risk andofobtain
portfolio,
the one
fundcould minimize r
contrast to other mathematical tools such as statistical allocation by employing the quadratic programming
the fund allocation by employing the quadratic programming (QP) proposed by M
models, forecasting, and simulation, mathematical (QP) proposedn by m Markowitz. The main objective
programming models allow the decision maker to find of QP ismin
to reduce
main objective of QP is to reduce ∑∑
σ ij variance,
the xthe
i x j variance,
which considers
which considersthe the upside and d
the optimal solution (Mokhtar et al., 2014). upside and downside
=i 1 =j 1 parts. The QP model is defined
Furthermore, the research community has focuses as follows.
The QP model is definednas follows. n
on the identification of efficient and effective solution
n m
s.t. E x
i i∑ ≥ L , = ∑
xi 1, 0 ≤ xi ≤ p= i, i 1,.
approaches for solving the mathematical programming min ∑∑ σ ij xi x j
i =1 i =1
(2)
model of portfolio optimization problems. Ordinarily, =i 1 =j 1

the methods can be categorized into two main types: where xin is the amount n
of invested funds in the financial a
s.t. ∑ Ei xi ≥ L, ∑ = xi 1, 0 ≤ xi ≤ p=
i, i 1,...., n
exact approach and heuristics approach. i =1 i =1

Exact Approaches where revenue

xi is the or amount the ofgrowth funds inpthe
investedfactor, i is financial
the highestassets level of the
of the firm i, L
where xi is the amount of invested funds in the financial
Numerous exact approaches have been suggested
assets of the firm i, L is the rate of revenue or the growth
to solve portfolio optimization either for single revenue or theorgrowth factor,ofpi the
sharesfactor, pthe bonds
is the highest is thefirm
level highest
of Elevel
thei,relative of the
i is the
relative investment
revenue
investment of the finaal
objective or multi-objective. Mixed integer quadratic i
allocated to the shares or the bonds of the firm i, Ei is
programming (MIQP) and QP are most widely used shares orthe the revenue
bonds of of thethefirm i, Ei is the revenue
financial asset of theof the financial
period studied,asset of the perio
to represent real-world problems in a deterministic σ ij is the common covariance of the revenue of the financial a
and sij is the common covariance of the revenue of the
manner. Nevertheless, another deterministic method σ ij is the common
financial covariance
asset i with of the revenue of the financial asset i with financial as
financial asset j. Note that for
of linear programming (LP) is employed but still goes for i = j , σ = σ 2 refers to the variance or the diagonal elem
ii i refers to the variance or the diagonal
through the transformation process from QP basedfor on i = j , σ ii = σ i2 refers to the variance or the diagonal elements of covariance. H
elements of covariance. He used QP for selecting the
mean absolute deviation (MAD). Papahristodoulou and
optimum
selecting theportfolio
optimum of the Malaysian Stock Exchange
Dotzauer (2004) formulated two different LP models selecting the optimum portfolio ofportfolio
the Malaysian of theStockMalaysian
Exchange byStock
dealingExch
the 10
by dealing the 10 biggest firms in 2014. The results
by transformation procedure, based on minimization of
in 2014. Theshowedresultsthat the optimum
showed portfolio
that the optimum includes
portfolio 22%22%
includes of of Axiata Grou
MAD and maximizes the minimum return (Maximin) in 2014. The results showed that
Axiata Group shares, 11% of Genting shares, 30% of the optimum portfolio includ
formulation. These models were then compared to of theGenting shares, 30% of Petronas Chemicals shares, 1% of Sime Darbi share
Petronas Chemicals shares, 1% of Sime Darbi shares,
classical quadratic programming formulation to test of Genting shares, 30% of Petronas Chemicals shares, 1%
and 36% of Tenaga Nasional shares.
to what extent all these formulations provide similar Tenaga Nasional shares.
Another quadratic programming model for the
portfolios. The results from this study showed that
Tenaga Nasional
portfolio shares.is done by Kulali (2016). He
optimization
the Maximin formulation yields the highest return Another quadratic programming model for the portfolio optimization is
tested Markowitz’s mean-variance approach to 252
and risk while the quadratic formulation provides the
(2016). He days of Markowitz’s
Another
tested data belonging
quadratic a year of approach
programming
mean-variance 2015 ontoIstanbul
model
252 daysfor the belon
of data por
lowest risk. In addition, all the three formulations were
Stock Exchange (BIST). The author compared two
found to outperform the top equity fund portfolios2015 in on Istanbul Stock Exchange (BIST). The author compared two hypothetica
hypothetical portfolios by considering diversification
Sweden and performed much better than the market (2016). He tested Markowitz’s mean-variance approach to 2
strategy: (a) 10 stocks with equal weights chosen from
portfolio (Mokhtar et al., 2014). considering diversification strategy: (a) 10 (b)
stocks withstocks
equal weights
three different industries and eight with chosen from
Another linear programming model for the 2015 on Istanbul Stock Exchange (BIST). The author comp
different weights. By employing Excel data solver, the
industries and (b) eight stocks with different weights. By employing Excel da
portfolio optimization is offered by Ibrahim, Kamil,
empirical results exhibited that the second hypothetical
and Mustafa (2008). In this study, the problem
considering
empirical provides more
results exhibiteddiversification
return than
that the strategy:
a portfolio
second with
hypothetical (a)equal
10 stocksmorewith
shares
provides returnequal
than a
was modelled as a mean-risk bi-criteria portfolio
of 10 stocks.
optimization problem with the mean absolute negative equal shares Bestof 10 and stocks.Kale (2000) also employed QP to solve
deviation of annual return from the average annual industries and (b) eight stocks with different weights. By
large-scale portfolio optimization. They proposed
return is used as the downside risk. In order to
the specialization of the QP algorithm for large-scale
evaluate the performance of the proposed model, the empirical results exhibited that the20 second hypothetical provi
portfolio optimization consists of three innovations:

equal shares of 10 stocks.

 also proved the convergence of the smallestCesaroneindex ruleet for
al. convex QP
(2009) obs
Cesarone et al. (2009) observed QP solve the Limited Asset Markowitz (LAM
Cesarone et al. (2009) observed QP solve
the classical the Limited
Markowitz modelAsset
tha
the classical Markowitz model that can be solved using QP, the LAM model is modeled
A Review on Shariah Stock Portfolio Optimization the classical Markowitz model that can be solved using151 QP,becoming
the LAMam
binary variables, thus
binary variables, thus becoming a mixed-integer quadratic programming (MIQP) probl
(a) determination of good starting point forbinary the variables, thus abecoming
thus becoming a mixed-integer
mixed-integer considerably quadratic
more
quadratic programmingdifficultprogramm
to solv
QP algorithm, (b) efficient considerably
solution of themore difficult(MIQP)
Karush- to solve. The LAM
problem that isisconsiderably
obtained asmore
follow:
difficult to
considerably
Kuhn-Tucker system for the active constraints using solve.more difficult
The LAM to solve.asThe
is obtained LAM is obtained
follow: n m as follow:
min ∑∑ σ ij xi x j
a small kernel matrix, and (c) handling of all n m
upper
min ∑∑ σ ij xi x j n m
=i 1 =j 1
and lower bounds as well as breakpoints for variable min ∑∑ σ ij xi x j n n
=i 1 =j 1 (3)
transactions costs implicitly rather than explicitly.
n n =i 1 =j 1
s.t. ∑ µ x
i i = ρ and ∑ xi =
The results produce improvements of ∑ µi1000
s.t. over xi = ρ and ∑ xi = 1 n
i =1 ∑ µ x = ρ and ∑ x = 1
n
=
i =1

xi 0 or li ≤ xi ≤ ui ,=
i =1

i
times in execution times for the optimizationi =1of large s.t. i i i

portfolios. x
= i 0 or li ≤ xi ≤ u i , i =i1 1,...., n,
= | supp(
i =1 x ) | ≤ K

xi 0 or li ≤ xi ≤ ui ,=
= i 1,...., n, | supp( x) | ≤ K
In Zhang and Zhang (2011), the authors built a where n is the number of assets, µ
series of pivoting algorithmwhere n is the
for solving number
convex QP of assets, µi is the expected return of asset i, σ ij is the covariance
and then shows how to use it together with a parameter where n is the number
where of assets,
n is the number µi ismithe
of assets, expected
is the expectedreturn return of asset i, σ
of asset i and asset j for i, j = 1,...., n
technique to solve the system of linear inequalities, of asset i, sij is the covariance of return of asset i and
of asset i and asset j for i, jasset = 1,...., n , ρ is the required level of return for the portfolio, th
j for i,j j=1,...,n,
convex quadratic programming, and mean-variance of asset i and asset for i, j =r1,....,
is the
n , required
ρ is the level
constraint requiredof returnlevelno
| supp( x) | is
of more
returntha fo
portfolio selection problems proposed by Markowitz. for the portfolio, the realistic constraint is
These algorithms are briefconstraint
for understanding is no more
| supp( x) | and no more
constraintquantity
than than
| supp( xx) | of
K Kassets
assetsfor
is each
no more
for supp(
than
= x) {i : xi > 0},,and
K assets for
andthethe quantity
supp(= x)a {i : xi > 0} , a
efficient for computing, as shown by the numerical i
asset should be limited within
examples and computerassetexperiments
should be for
limited
1072 within
stocks. a given giveninterval
interval [li , ui ] . He
He established
establishedthethe new new method so-called
method
They also proved the convergence of the smallest index so-called reduction to standard QP. It tested on five
rule for convex QP. reduction to standard QP. It tested on newfivedata newsetsdatainvolving 21 real-world capital market
sets involving real-world capital market indices
21
Cesarone et al. (2009) observed QP solve the indices from major stock markets that have been
Limited Asset Markowitz (LAM). Unlike the classical used in Chang et al. (2000).
from major stock markets that have been used in Chang et al. (2000). The performance The performance of the of the
Markowitz model that can be solved using QP, the portfolios obtained from the LAM model is more
LAM model is modeled by adding
portfolios binaryfrom
obtained the LAMefficient
variables, model compare is more to the classical
efficient Markowitz
compare to theportfolio.
classical Markowitz

portfolio.
Table 7. Solution of Exact Approaches for Portfolio Optimization
Table 7
Exact approaches
Authors Data analyzed
Solution of Exact ApproachesLPfor MAD
Portfolio QP PA
Optimization MIQP
Papahristodoulou Monthly returns from 67 shares traded in the  
& Dotzauer, 2004 Stockholm Stock Exchange (SSE), between January Exact approaches
1997 and December 2000
Authors Data analyzed
Ibrahim et al., 2008 60 random stocks registered on the main board of   LP MAD QP PA MIQP
Bursa Malaysia and listed on May 2004
Yousfat, 2015 10 biggest firms posted on Bursa Malaysia during 
Papahristodoulo
2014
Monthly returns from 67 shares traded in  
Kulali, 2016 252 days of data belonging a year of 2015 on Istanbul 
uStock
& Dotzauer,
Exchange (BIST)the Stockholm Stock Exchange (SSE),
Best & Kale, 2000 Large-scale portfolio optimization 
2004 between January 1997 and December
Zhang & Zhang, Arbitrary weekend closed prices of 1072 stocks. 
2011
2000
Cesarone et al., Weakly price data from March 1992 to September 
2009 1997 for the Hang Seng, DAX, FTSE 100, S&P 100,
Ibrahim et 225
and Nikkei al., capital
60market
random stocks
indices. registered
In addition, on the main  
also around 2000 assets taken from the OR-Library.
2008 board of Bursa Malaysia and listed on
Notes: LP = Linear Programming, MAD = Mean Absolute Deviation, QP = Quadratic Programming, MIQP = Mixed Integer
Quadratic Programming, PA = Pivoting. Algorithm and missing of Shariah stock cases for the rare be obtained.
May 2004

Yousfat, 2015 10 biggest firms posted on Bursa 

Malaysia during 2014

152
Heuristic Approaches
Heuristic is a popular non-deterministic
approach for solving hard constrained problems.
It attempts to yield a good and fast approximation
to an optimal solution, that is, seeking approximate
answers to the solution obtained from the exact
approach. Requirements for mathematical
sophistication or programming skills make this
approach flexible because adding, removing,
or changing objective functions or constraints
can effortlessly be achieved (Gilli & Schumann,
2012). There are many studies applying heuristic
approaches to solve the problem of portfolio
optimization. Crama and Schyns (2003) applied
heuristic technique based on Simulated Annealing
(SA) to an extended version of the mean-variance
model with trading and turnover constraints.
Such model is contained within Mixed Integer
Quadratic Programming which is then applied
to solve a complex portfolio selection model.
Computational results for problems with up to 151
stocks seem to show that the approach is promising
for medium size problems. Computational results
for problems with up to 151 stocks seem to show
that the approach is promising for medium size
problems (Mokhtar et al., 2014).
John (2014) established five hill-climbing
algorithms, namely HC-S, HC-S-R, HC-C,
HC-C-R, and guided local search (GLS). Hill
climbing, as one heuristic method, is applied to
approximate the solution of a problem. It is first
tested to standard portfolio optimization problem
by retaining Markowitz’s constraints that the
investor has a fixed budget and no short-selling.
The algorithms are next applied to the extended
portfolio optimization problem by considering
cardinality constraints. Results are benchmarked
with the threshold accepting (TA) algorithm and
QP. The finding suggests that such five algorithms
developed have a similar solution to the exact
solution of QP and surprisingly outperformed TA
in the extended portfolio optimization.
Chang et al. (2009) introduced a heuristic
approach based on genetic algorithm (GA)
for solving portfolio optimization problems
N. S. M. Mussafi & Z. Ismail
in different risk measures and compared its
performance to mean-variance model in cardinality
constrained efficient frontier. The authors collected
three different risk measures based upon mean-
variance by Markowitz, semi-variance, mean
absolute deviation, and variance with skewness.
They showed that the problems could be solved
effectively by GA if mean-variance, semi-
variance, mean absolute deviation, and variance
with skewness were used as the measures of
risk. They conducted empirical tests to prove the
robustness of their heuristic method by verifying
the three data sets collected from main financial
markets.
Kapiamba et al. (2015) compared a bi-criteria
problem having two conflicting criteria to optimize
portfolio selection simultaneously. It is well known
that such a combinatorial problem is intractable
with exact methods for large dimension problems.
Then meta-heuristics are useful to find a good
approximation of the efficient set. He evaluated
the efficiency of two heuristic methods, namely
the simulated annealing and the genetic algorithm,
to solve the portfolio selection problem. Results
of the study indicated that, in terms of calculation
time, simulated annealing appears more efficient
than the genetic algorithm.
Busetti (2006) developed a realistic portfolio
optimization model. He then investigates the efficiency
of its solution by two heuristic methods, that is, GA
and tabu search (TS). The model is based on the
classical mean-variance approach, enhanced with
floor and ceiling constraints, cardinality constraints,
and nonlinear transaction costs that include a
substantial illiquidity premium, and is applied to a
large stock portfolio. The solution is benchmarked
with the insights provided by the optimization of real
portfolios. The study showed that for large portfolios,
the performance of genetic algorithms is three orders
of magnitude better than that of TS.
Another interesting solution approach was
presented by Fernández and Gómez (2007), who
employed Hopfield neural networks to solve the
mean-variance model with cardinality and bounding
constraints. The authors also compared the approach
with GA, TS as well as SA and performed
the computational experiments using five sets of
A Review on Shariah Stock Portfolio Optimization 153

Table 8. Solution of Heuristic Approaches for Portfolio Optimization

Authors
Crama & Schyns, 2003
John, 2014
Chang et al., 2009
Kapiamba et al., 2015
Busetti, 2006
Fernández & Gómez, 2007
Heuristic approaches
Data analyzed
SA HC GA TS
Weekly prices of 151 U.S. stocks covering different 
traditional sectors for 484 weeks, from 6 January
1988 to 9 April 1997, extracted from the DataStream
database.
230 assets are from DAX stock exchange. The data 
used were daily returns over 1001 days.
Historical daily data collected in the HANG SENG, 
FTSE and S&P 100 with price data of 33, 93, and 99
assets respectively from January 2004 to December
2006.
Artificial data  
Arbitrary 100 largest stocks by market capitalization  
in the world.
Weekly prices listed from Hang Seng in Hong Kong,   
DAX 100 in Germany, FTSE 100 in the U.K., S&P
100 in USA and Nikkei 225 in Japan with price data
of 31, 85, 89, 98 and 225 assets respectively from
March 1992 to September 1997.

Notes: SA = Simulated Annealing, HC = Hill-Climbing, GA = Genetic Algorithm, TS = Tabu Search

benchmark data that have been used in T. J. Chang
et al. (2000). Although the results showed that none
of the four has clearly outperformed the others, when
dealing with problem demanding portfolios with low
investment risk, the proposed method provides better
solutions than the other heuristics (Mokhtar et al.,
2014). There are several works not mentioned that
have already taken steps in the direction of what the
authors called the Research Gap section.
Research Gap
There are some missing elements in the discussion
on stock portfolio optimization. First, the exploration
of Shariah stock market is still limited. Ho et al.
(2014) has exposed that the increase on the number
of global Shariah stock indices in the world is
evidence of relatively large growth potential and
profitability. Moreover, some scholars also claimed
that the performance of the Shariah stock market has
a tendency to be preferable with regard to return and
stable in defiance of financial crisis compared to the
conventional ones (Abu Bakar & Ali, 2014; Karim et
al., 2014; Reddy & Fu, 2014). In more detail, Lusyana
and Sherif (2017) said that Indonesia, with the largest
Muslim population in the world, was favorable by reason
of the increasing number of investors who are concerned
in the ethical investment. Indeed, the quantity of these
works, which mainly refer to the Islamic stock market,
is still less frequent than the research results related to
data from conventional platforms.
Second, the gap in methodology, particularly on
how to select the promising portfolio. The study on
blending between fundamental and technical analysis is
still lacking and has never even been discussed. Many
researchers stick just to one of either fundamental or
technical. Some believed that fundamental analysis is
the only proper instrument that can comprehensively
capture the intrinsic value of a company (Jakpar et al.,
2019; Mustilli et al., 2018; Silva et al., 2014; Suresh,
2013). On the contrary, others claimed that technical
analysis is more powerful because it can predict future
risks or returns that are not yet known on the day of
the investment decision (AlHalaseh et al., 2016; Kulali,
2016; Lee & Eid, 2018; Mangram, 2013; Rockafellar
& Uryasev, 2016).
154
Third, the quadratic programming as a standard
form in portfolio optimization still has drawback and
require some form of modification or improvement for
solving stock portfolio problem. Researchers have tried
to modify quadratic programming with their respective
findings, which have proven difficult to complete
large-scale portfolio optimization (Best & Kale, 2000;
Cesarone et al., 2013; Zhang & Zhang, 2011).
Finally, another gap in methodology, especially
on portfolio optimization. The comparative study
between exact and approximation methods, that is,
standard quadratic programming, modified quadratic
programming, and heuristic approaches has never
been studied. In handling portfolio optimization, some
scientists only focus on the heuristic approach (Chang
et al., 2009; Crama & Schyns, 2003; Kapiamba et al.,
2015), while others concentrate on the exact approach
(Ibrahim et al., 2008; Kulali, 2016; Yousfat, 2015).
Conclusion
Stock investment has become an interesting
topic not only in economics but also in mathematics.
This review provided the current situation on stock
investment, particularly the stock market, along with
its dynamics. It is also considered a new trend of
investments based on ethical so-called Shariah stock
investment. Such a concept has grown to become
one of the world’s profitable economic forces so that
many well-known stock markets in the U.S., Europe,
and Asia are involved in this investment. Moreover,
Shariah stock has also been proven to be better than
conventional, especially related to the rate of return and
stability in a crisis. Nevertheless, the main drawback
of investing in Shariah stocks is the limited options
due to the screening of trade mechanisms and halal
products. Investors in conventional stocks have so
many options that they can tailor their portfolios to
meet any investment objective, whereas investors in
Shariah stocks have fewer options to pick from.
The review highlighted techniques on stock
portfolio selection, that is, technical analysis and
fundamental analysis. By exploring each of the
advantages and disadvantages of the two analyzes
based on previous research, the stock portfolio
assessment can be evaluated better. In view of that,
the stock portfolio selection might be executed by
considering the properties on fundamental analysis as
a base choice, but it seems that it needs to be verified
N. S. M. Mussafi & Z. Ismail
by technical analysis. Therefore the combination
of two altered methods would rather give a more
comprehensive decision.
The findings in this review shown that stock
portfolio optimization eventually turns into an
alternative tool for both theorists and practitioners in
dealing with a financial decision. It deliberated two
common approaches used by researchers. The exact
approach seems to be the popular method for solving
stock portfolio optimization rooted in Markowitz
properties. However, based on the review, the heuristic
approach could approximate the solution of problems
as an exact solution for a large scale of portfolios and
has less execution time. Therefore, this review led to
the further development of stock portfolio optimization
by comparing the models of the two approaches and
their results in the real case of stock market.
References
Abu Bakar, M. A. A., & Ali, N. (2014). Performance
measurement analysis: Shariah-compliant vs. non
Shariah-compliant securities. Malaysian Accounting
Review, 13(1), 75–108.
Alam, M. M., Akbar, C. S., Shahriar, S. M., & Elahi, M. M.
(2017). The Islamic Shariah principles for investment
in stock market. Qualitative Research in Financial
Markets, 9(2), 132–146.
Albaity, M., & Rubi, A. (2008). Performance of Syariah
and composite indices: Evidence from Bursa Malaysia.
Asian Academy of Management Journal of Accounting
and Finance, 4(1), 23–43.
AlHalaseh, R. H. S., Islam, M. A., & Bakar, R. (2016).
Portfolio selection problem: Models review. The Social
Sciences, 11(14), 3408–3417.
Andersen, T. J. (2008). The performance relationship of
effective risk management: Exploring the firm-specific
investment rationale. Long Range Planning, 41(2),
155–176.
Best, M., & Kale, J. (2000). Quadratic programming for
large-scale portfolio optimization. In Financial Services
Information Systems (pp. 513-529). CRC Press. https://
doi.org/10.1201/9780203997611
Busetti, F. (2006). Heuristic approaches to realistic portfolio
optimisation. WIT Transactions on Modelling and
Simulation, 43, 361–370. https://doi.org/10.2495/
CF060351
Campanella, F., Mustilli, M., & D’Angelo, E. (2015).
Efficient market hypothesis and fundamental analysis:
An empirical test in the European securities market.
Review of Economics and Finance, 6(1), 27–42.
Cesarone, F., Scozzari, A., & Tardella, F. (2009). Efficient
A Review on Shariah Stock Portfolio Optimization
algorithms for mean-variance portfolio optimization
with hard real-world constraints. W3.Uniroma1.It,
LXXII, 37–56. Retrieved from http://host.uniroma3.it/
docenti/cesarone/Store/CesaroneScozzariTardella2009-
EfficientAlgorithmsForMean-Variance.pdf%5Cnhttp://
w3.uniroma1.it/Tardella/CesaroneScozzariTardella
2009 - EFFICIENT ALGORITHMS FOR MEAN-
VARIANCE.pdf
Cesarone, F., Scozzari, A., & Tardella, F. (2013). A new
method for mean-variance portfolio optimization with
cardinality constraints. Annals of Operations Research,
205(1), 213–234. https://doi.org/10.1007/s10479-012-
1165-7
Chang, T. J., Meade, N., Beasley, J. E., & Sharaiha, Y. M.
(2000). Heuristics for cardinality constrained portfolio
optimisation. Computers and Operations Research,
27(13), 1271–1302. https://doi.org/10.1016/S0305-
0548(99)00074-X
Chang, T. J., Yang, S. C., & Chang, K. J. (2009). Portfolio
optimization problems in different risk measures using
genetic algorithm. Expert Systems with Applications,
36(7), 10529–10537. https://doi.org/10.1016/j.
eswa.2009.02.062
Cheng, P. (2001). Comparing downside-risk and mean-
variance analysis using bootstrap simulation. Journal
of Real Estate Portfolio Management, 7(3), 225–238.
Cohen, G., Kudryavtsev, A., & Hon-Snir, S. (2011).
Stock market analysis in practice: Is it technical or
fundamental? Journal of Applied Finance & Banking,
1(3), 125–138.
Crama, Y., & Schyns, M. (2003). Simulated annealing for
complex portfolio selection problems. European Journal
of Operational Research, 150(3), 546–571. https://doi.
org/10.1016/S0377-2217(02)00784-1
Dania, A., & Malhotra, D. K. (2013). An empirical
examination of the dynamic linkages of faith-based
socially responsible investing. The Journal of Wealth
Management, 16(1), 65–79. https://doi.org/10.3905/
jwm.2013.16.1.065
de Souza, M. J. S., Ramos, D. G. F., Pena, M. G., Sobreiro, V.
A., & Kimura, H. (2018). Examination of the profitability
of technical analysis based on moving average strategies
in BRICS. Financial Innovation, 4(1). https://doi.
org/10.1186/s40854-018-0087-z
Drakopoulou, V. (2016). A review of fundamental and
technical stock analysis techniques. Journal of Stock &
Forex Trading, 05(01), 1–8.
Duong, M. H. & Siliverstovs, B. (2006). The stock market
and investment [Unpublished manuscript].
Eiamkanitchat, N., Moontuy, T., & Ramingwong, S. (2017).
Fundamental analysis and technical analysis integrated
system for stock filtration. Cluster Computing, 20(1),
883–894. https://doi.org/10.1007/s10586-016-0694-2
Fernández, A., & Gómez, S. (2007). Portfolio selection
155
using neural networks. Computers and Operations
Research, 34(4), 1177–1191. https://doi.org/10.1016/j.
cor.2005.06.017
Fu, Y. (2019). On generalizations of mean-variance model.
Journal of Physics: Conference Series, 1168(5). https://
doi.org/10.1088/1742-6596/1168/5/052011
Geambasu, C., Sova, R., Jianu, I., & Geambasu, L. (2013).
Risk measurement in post-modern portfolio theory:
Differences from modern portfolio theory. Computation
& Economic Cybernetics Studies & Research, 47(1),
113–132.
Geumei, D. (2018). The Islamic capital markets versus
the conventional capital markets. American Journal of
Humanities & Islamic Studies, 1(1), 1–8.
Ghaeli, M. R. (2017). Price-to-earnings ratio: A state-of-
art review. Accounting, 3(2), 131–136. https://doi.
org/10.5267/j.ac.2016.7.002
Gilli, M., & Schumann, E. (2012). Heuristic optimisation
in financial modelling. Annals of Operations Research,
193(1), 129–158. https://doi.org/10.1007/s10479-011-
0862-y
Ho, C. S. F., Abd Rahman, N. A., Yusuf, N. H. M., &
Zamzamin, Z. (2014). Performance of global Islamic
versus conventional share indices: International
evidence. Pacific Basin Finance Journal, 28, 110–121.
https://doi.org/10.1016/j.pacfin.2013.09.002
Hooke, J. (2010). Security analysis and business
valuation on Wall Street (2nd ed.). John Wiley &
Sons, Inc.
Hussain, M., Shahmoradi, A., & Turk, R. (2015). An
overview of Islamic finance. International Monetary
Fund.
Ibrahim, K., Kamil, A. A., & Mustafa, A. (2008).
Portfolio selection problem with maximum downside
deviation measure: A stochastic programming approach.
International Journal of Mathematical Models and
Methods in Applied Sciences, 2(1), 124–129.
Jakpar, S., Tinggi, M., Tak, A. H., & Chong, W. Y. (2019).
Fundamental analysis vs technical analysis: The
comparison of two analysis in Malaysia stock market.
UNIMAS Review of Accounting and Finance, 1(1),
38–61.
John, C. (2014). Portfolio Optimization by Heuristic
Algorithms (Doctoral dissertation, University of Essex).
1-92.
Kapiamba, J. N., Mubenga, P. K., & Ulungu, B. (2015).
Simulated annealing vs genetic algorithm to portfolio
selection. International Journal of Scientific and
Innovative Mathematical Research, 3(5), 18–30.
Karim, B. A., Datip, E., & Shukri, M. H. M. (2014).
Islamic stock market versus conventional stock market.
International Journal of Economics, Commerce and
Management, 2(11), 1–9.
Kearney, C., & Daly, K. (1998). The causes of stock market
156
volatility in Australia. Applied Financial Economics,
8(6X), 597–605.
Kolbadi, P., & Ahmadinia, H. (2011). Examining Sharp,
Sortino and Sterling ratios in portfolio management,
evidence from Tehran Stock Exchange. International
Journal of Business and Management, 6(4), 222-236.
Kulali, I. (2016). Portfolio optimization analysis with
Markowitz quadratic. European Journal of Business and
Management, 8(7), 73–79.
Lee, S. C., & Eid, W., Jr. (2018). Portfolio construction
and risk management: Theory versus practice. RAUSP
Management Journal, 53(3), 345–365.
Louche, C., Arenas, D., & van Cranenburgh, K. C. (2012).
From Preaching to Investing: Attitudes of Religious
Organisations Towards Responsible Investment.
Journal of Business Ethics, 110(3), 301–320. https://doi.
org/10.1007/s10551-011-1155-8
Lusyana, D., & Sherif, M. (2017). Shariah-compliant
investments and stock returns: Evidence from the
Indonesian stock market. Journal of Islamic Accounting
and Business Research, 8(2), 143–160.
Mangram, M. E. (2013). A simplified perspective of the
Markowitz. Global Journal of Business Research, 7(1),
59–70.
Markowitz, H. (1952). Portfolio Selection. The
Journal of Finance, 7(1), 77–91. https://doi.
org/10.1111/j.1540-6261.1952.tb01525.x
Mohammed, N. F., Puat, S. A., Amirrudin, M. S.,
& Hashim, A. (2020). Leverage, liquidity and
profitability ratios: Accountability of Malaysian
listed oil and gas firms. Humanities & Social Sciences
Reviews, 8(2), 941–947. https://doi.org/10.18510/
hssr.2020.82104
Mokhtar, M., Shuib, A., & Mohamad, D. (2014).
Mathematical programming models for portfolio
optimization problem: A review. International Journal
of Social, Education, Economics and Management
Engineering, 8(2), 428–435.
Moosa, I., & Li, L. (2011). Technical and fundamental
trading in the Chinese stock market: Evidence based on
time-series and panel data. Emerging Markets Finance
and Trade, 47(sup1), 23–31.
Mussafi, N. S. M. (2012). Optimisasi portofolio resiko
menggunakan model Markowitz Mvo [Risk Portfolio
Optimization using Markowitz MVO Model].
AdMathEdu: Jurnal Ilmiah Pendidikan Matematika,
Ilmu Matematika Dan Matematika Terapan, 1(1), 1–8.
https://doi.org/10.12928/admathedu.v1i1.4879
Mustilli, M., Campanella, F., & D’Angelo, E. (2018).
Abnormal returns and fundamental analysis in
institutional Investors’ decision-making: An agency
theory approach. International Business Research,
11(2), 55–69.
Neely, C. J., Rapach, D. E., Tu, J., & Zhou, G. (2010). Out-
N. S. M. Mussafi & Z. Ismail
of-sample equity premium prediction: Fundamental
vs. technical analysis. Research Collection Lee Kong
Chian School of Business. Retrieved from https://www.
business.unsw.edu.au/About-Site/Schools-Site/banking-
finance-site/Documents/G.%20Zhou%20-%20Out-of-
Sample%20Equity%20Premium%20Prediction.pdf
Osmani, N. M., & Abdullah, F. (2009). Towards an
Islamic stock market: A review of classical and modern
literatures. International Review of Business Research
Paper, 5(5), 121-130.
Papahristodoulou, C., & Dotzauer, E. (2004). Optimal
portfolios using linear programming models. Journal of
the Operational Research Society, 55(11), 1169–1177.
https://doi.org/10.1057/palgrave.jors.2601765
Pentheny, G. (2009). Analysis of stock market investment
strategies [Unpublished bachelor’s thesis]. Worcester
Polytechnic Institute.
Pilotte, E. A., & Sterbenz, F. P. (2006). Sharpe and Treynor
ratios on Treasury Bonds*. The Journal of Business,
79(1), 149–180.
Rasiah, D. (2012). Post-modern portfolio theory supports
diversification in an investment portfolio to measure
investment’s performance. Journal of Finance and
Investment Analysis, 1(1), 69–91.
Reddy, K., & Fu, M. (2014). Does Shariah compliant
stocks perform better than the conventional stocks? A
comparative study stocks listed on the Australian Stock
Exchange. Asian Journal of Finance & Accounting,
6(2), 155-170.
Ridha, M. B., & Alnaji, L. (2013). The methods of
measurement and analysis of risks in businesses: A case
study on the Jordan Valley authority. Developing Country
Studies, 3(2), 60–72.
Rockafellar, R. T., & Uryasev, S. (2016). Optimization of
conditional value-at-risk. The Journal of Risk, 2(3),
21–41.
Rom, B. M., & Ferguson, K. W. (2009). Post-modern
portfolio theory comes of age. The Journal of Investing,
2(4), 27–33.
Roman, D., & Mitra, G. (2009). Portfolio selection models: A
review and new directions. Wilmott Journal, 1(2), 69–85.
Sabol, A., Šander, M., & Fučkan, D. (2013). The concept of
industry life cycle and development of business strategies.
In International conference “Active Citizenship by
Management, Knowledge Management & Innovation
Knowledge and Learning”, (pp. 635-642). Zadar.
Samra, Y. M., & Joseph, G. (2018). Is there that much of
a difference: A comparison between conventional and
Islamic investment vehicles. Journal of Internet Banking
and Commerce, 23(2), 1–11.
Sarykalin, S., Serraino, G., & Uryasev, S. (2008). Value-
at-risk vs. conditional value-at-risk in risk management
and optimization. In State-of-the-art decision-making
tools in the information-intensive age (pp. 270-294).
A Review on Shariah Stock Portfolio Optimization
Informs.
Sathyanarayana, S., & Gargesa, S. (2018). An analytical
study of the effect of inflation on stock market returns.
IRA-International Journal of Management & Social
Sciences, 13(2), 48–64xx. https://doi.org/10.21013/
jmss.v13.n2.p3
Sharpe, W. F. (1994). The Sharpe ratio. The Journal of
Portfolio Management, 21(1), 49–58.
Siegel, J. J., & Coxe, D. G. (2002). Stocks for the long run
(3rd ed.). McGraw-Hill.
Silva, A., Neves, R., & Horta, N. (2014). Portfolio
optimization using fundamental indicators based on
multi-objective EA. In IEEE/IAFE Conference on
Computational Intelligence for Financial Engineering,
Proceedings (pp.158–165). IEEE.
Suresh, A. S. (2013). A study on fundamental and technical
analysis. International Journal of Marketing, Financial
Services & Management Research, 2(5), 44–59.
Thomsett, M. C. (2005). Getting started in fundamental
analysis. John Wiley & Sons, Inc.
Todoni, M.-D. (2015). A post-modern portfolio management
approach on CEE markets. Procedia Economics and
Finance, 32(15), 1362–1376. https://doi.org/10.1016/
s2212-5671(15)01513-0
Treynor, J.L. (1965) How to Rate Management of Investment
157
Funds. Harvard Business Review, 43, 63-75.
Vincent, O., & Bamiro, K. (2013). Fluctuations in stock
market prices: What went wrong, its implications to
Nigerian economy? International Journal of Computer
Applications, 63(19), 13–20.
Wafi, A. S., Hassan, H., & Mabrouk, A. (2015). Fundamental
analysis vs technical analysis in the Egyptian stock
exchange–Empirical study. International Journal of
Business and Management Study–IJBMS, 2(2), 212-218.
Yadav, S. (2017). Stock market volatility - A Study of Indian
Stock Market. Global Journal for Research Analysis,
6(4), 629–632.
Yousfat, A. (2015). The portfolio selection by using
quadratic programming approach case study of Malaysia
stock exchange. International Journal of Engineering
and Technology, 7(4), 1361–1369.
Yusof, R., & Majid, M. (2009). Stock market volatility
transmission in Malaysia: Islamic versus conventional
stock market. Journal of King Abdulaziz University-
Islamic Economics, 20(2), 19–40.
Zhang, Z., & Zhang, H. (2011). An efficient and concise
algorithm for convex quadratic programming and its
application to Markowitz’s portfolio selection model.
Technology and Investment, 02(04), 229–239. https://
doi.org/10.4236/ti.2011.24024