TEORI PASAR MODAL DAN INVESTASI

CRITICAL REVIEW JURNAL SAP 7

Selection of Orthogonal Investment Portofolio
using Evolino RNN Trading Model

OLEH. KELOMPOK 4
Magister Maksi Angkatan XXIV Kelas B:

Ida Bagus Nyoman Ramartha Putra (1981611033 / 02)

Ni Wayan Trisna Kusumayanti (1981611039 / 08)
A.A. Rai Niti Darmika Sukawati (1981611041 / 10)
Made Ayu Prilla Winda Puspita (1981611053 / 22)
Ni Made Madani Hapsari (1981611054 / 23)

PROGRAM STUDI MAGISTER AKUNTANSI

FAKULTAS EKONOMI DAN BISNIS UNIVERSITAS UDAYANA
DENPASAR
2020
Kalimat pertama menjorok ke dalam. Spsasi 1.15, times new roman, size
12. Perhatikan terjemahan agar logis dan tata penulisannya dirapikan.
Selection of Orthogonal Investment Portofolio using
1 Judul Evolino RNN Trading Model
Pemilihan Portofolio Investasi Orthogonal menggunakan
Model Perdagangan Evolino RNN
Elsevier: Social and Behavioral Science 110 (2014) 1158-
2 Sumber
1165
3 Penulis Nijole Maknickiene
Investing in financial market require the reliable
predicting of expecting returns, assessment of risk and
reliability. Principle of portfolio orthogonality was using to
reduce the risk of the investment. An artificial intelligence
4 Latar Belakang
system may reveal new opportunities for using this
principle. Prediction of recurrent neural networks ensemble
is stochastically informative distribution, which is helpful
for portfolio selection.
Investment decisions are always made under uncertainty;
it is possible only in a heuristic approach to the maximum
result. Potential distributions best reflect the investment
5 Tujuan Penelitian opportunities. The aim of paper is to select investment
portfolio using parameters of potential distribution obtained
by the model based on ensemble of recurrentneural
networks.
6 Kajian Teoritis

 A  a
7 Kajian Empiris  A  a
 a  a
8 Sampel Penelitian

It is known, that ensembles can predict more accurate,

then single RNN. Authors constructed Evolino RNN based
prediction model, ensemble from 176 RNN. Influence of
9 Metode Penelitian number of RNN in ensemble was investigated by authors in
earlier work (Maknickiene & Maknickas, 2013).
+page 1162, data step, input step dan prediction step –
ringkas.
 Page 1162 setelah steps itu sampe sebelum conclusions
itu hasil. Bisa dipilah data nya yang penting masukin kesini.
Temuan dan/atau
10
Hasil Penelitian

Main portfolio selection principals of modern portfolio

theory- maximising expected return and minimising
riskiness – are successfully using together with artificial
intelligence. Potential distributions best reflects the
investment opportunities in conditions of uncertainty.
Model, using ensemble of Evolino RNN, is reliable
forecasting tool useful for investors in currency market. Its
Kesimpulan prediction is stochastically informative potential
11 distribution, which is helpful for portfolio selection. Shape
Penelitian
and parameters of distribution influence decision making in
currency market.

Testing in Forex market in real time confirmed, that

assessment of portfolio riskiness, finding most orthogona
elements of portfolio, influence better results for trading in
real market. Orthogonality and optimization influence the
better selection of investment portfolio.

1.
Keterbatasan
12
Penelitian