You are on page 1of 6

Universidad del Pacífico

Facultad de Economía y Finanzas


2020-II

Quantitative Finance
Final Project

Group Members:
Acuña, Magaly
Cotrina, Viviana
Flores, Álvaro
Ponte, Diana
Porras, Neil

Teaching Assistant:
Fernández, Gonzalo

2020
1. Preliminary back test
We downloaded the data from 2009 to the end of September 2020 using information
from Yahoo finance, at the same time we created a function with the log of returns for
the indicated date. Using the Risk Parity strategy, we create twelve-month size moving
windows with one-month steps in order to create the weights to be initially assigned on
each rebalancing date. Then we use the “CalculateRiskParity” function that shows us
the weights of a portfolio following the risk parity strategy of the covariance matrix of
the returns matrix found previously.

Figure Nº1: Project Risk Parity with SHY

As we can see in the figure Nº1, most of the portfolio is made up of SHY shares during
the proposed dates.
Tangency and Cvar strategies are calculated following the same steps mentioned above:
portfolio creation, window creation and weight calculation respectively. The figure Nº2
shows that the portfolio according to Tangency is still mainly made up of SHY and SPY
shares and during the period of December 2018 it is completely concentrated in EZU
shares.

Figure Nº2: Project Tangency with SHY


As shown in figure Nº3, CVaR's portfolio is very little diversified throughout the period
analyzed, it is made up almost entirely of SHY shares, so we can assume that it does not
assume much risk. As seen above, SHY's assets occupy more than 40% of portfolios.

Figure Nº3: Project CVaR with SHY

As seen above, SHY's assets occupy more than 40% of portfolios. For this reason, to
carry out the strategy we will initially consider only 12 assets, excluding SHY from the
beginning. The portfolio with risk parity will remain as shown in figure Nº4, without
participation of SHY shares, as we can see, the distribution of the shares within the
portfolio is more equitable.

Figure Nº4: Project CVaR without SHY


On the other hand, when comparing the three strategies mentioned, we observe an
upward trend in the Risk Parity Index, outperforming the other two in almost all
periods. Even the risk parity without SHY has a cumulative return advantage over the
other strategies as shown in figure Nº5.

Figure Nº5: Cumulative Return

Regarding the drawdowns, we observe that over time the 3 strategies have points in
common due to economic scenarios with negative impact. First of all, we can see an
approach to zero during 2010, probably associated with the recession due to the
technology bubble. During the period 2014-2015 another joint approach is seen due to
the collapse of crude oil prices. Another point in common occurred during 2019 due to
the trade war between the United States and China, in addition to the global economic
slowdown.

2. Base Model: Risk Parity, Momentum and Trend Following in ETF


Allocation
We have two types of ETFs:
 Equity ETF Basket: we have considered a high yield corporate bond ETF, HYG,
as an Equity ETF. We have the following nine ETFs in this basket: SPY, EEM,
QQQ, FXI, EWJ, EZU, AAXJ, ILF and HYG.
 Fixed-income ETF Basket: We have the following three ETFs in this basket:
LQD, TLT, IEF and SHY.
*Cash: We have considered SHY ETF as an approximation to cash, it will not be used
to compute risk parity weights.

Risk parity means that less volatile ETFs represents a larger portion of the portfolio than
more volatile ETFs. (Clare et al., 2016)

Momentum involve ranking ETFs based on their achieved returns and then buying the
best and selling the losers. This constitute a relative measure because, we choose the
best performing ETFs relative to other ETFs. In our case, we will not take short
positions; we will allocate the losers as SHY ETF if the following rule is true. (Faber,
2012)

Trend following propose a straightforward rule: allocate in the portfolio each ETF that
has a price at the close greater than 10-month moving average, otherwise allocate that
portion of the portfolio to cash, in our case, the SHY ETF.

The objective of this strategy is to achieve an improvement in risk-adjusted performance


of our portfolio compared to traditional buy and hold portfolio, i.e. to construct a
portfolio that has equity-level returns with bond-level volatility. (Faber 2010)

3. The strategy

Risk parity:
On the last trading day of each month, at the close, we have determined portfolio
weights based on simple 12-month Risk parity. We found that the weight corresponding
to the less volatile ETF, SHY, is concentrated in a larger proportion than other more
volatile ETFs, for this reason, we will determine the portfolio weights based 12-month
Risk parity for all ETFs in basket without including SHY.

Dual Momentum:

Momentum:

At the close on the last trading day of that month, we will calculate the one, three, six,
and twelve-month total returns for each of nine Equity ETFs. Then, we find the simple
average of those four returns for each ETF. The top five Equity ETFs with higher
returns (can be positive or negative) are candidates to maintain its weights calculated by
risk parity. We allocate the weights of the four Equity ETFs with lower simple average
return to SHY ETF.

Trend Following:

We go long on the top five ETFs unless those ETF will close below its 10-month
moving average, in which case its/their portion of the portfolio calculated by risk parity
is allocated to SHY.

We hold positions until the next month. We rebalance the entire portfolio monthly.
4. Bibliography

Clare, A., Seaton, J., Smith, P. N., & Thomas, S. (2016). The trend is our friend: Risk
parity, momentum and trend following in global asset allocation. Journal of
Behavioral and Experimental Finance, 9, 63–80.
https://doi.org/10.1016/j.jbef.2016.01.002

Faber, M. T. (2012). Relative Strength Strategies for Investing. SSRN Electronic


Journal, 90245(April). https://doi.org/10.2139/ssrn.1585517

Faber, M. T., & Paper, W. (2010). A Quantitative Approach to Tactical Asset


Allocation Cambria Investment Management has been managing investments for
individuals and institutions since 2007 . To learn more about all of our investment
offerings , please contact us for more information : P. PA - Book, May 2006, 1–70.

You might also like