THESIS. Forecasting Stock Price Index Using Artificial Neural. Networks in the Indonesian Stock Exchange

THESIS

Forecasting Stock Price Index Using Artificial Neural

Networks in the Indonesian Stock Exchange

SOUKKHY TIPHIMMALA Sdut.Id: 125001870/PS/MM

PROGRAM STUDY MASTER MANAGEMENT

PROGRAM GRADUATE

UNIVERSITY OF ATMA JAYA YOGYAKARTA

2014

ii INTISARI

Indeks harga saham adalah faktor yang signifikan mempengaruhi awal pada pengambilan keputusan keuangan investor. Itu sebabnya memprediksi gerakan yang tepat dari indeks harga saham jauh dianggap. Penelitian ini bertujuan untuk mengevaluasi efektivitas penggunaan indikator teknis, seperti A / D Oscillator, Moving Average, RSI, CCI, MACD, dll dalam memprediksi pergerakan Bursa Efek Indeks Harga Indonesia (BEI). Sebuah jaringan syaraf tiruan digunakan untuk peramalan indeks harga saham. Data yang ada dicapai dari Yahoo.Finance. Untuk menangkap hubungan antara indikator teknis dan tingkat indeks di pasar untuk periode diselidiki, jaringan saraf propagasi kembali digunakan. Kinerja statistik dan keuangan dari teknik ini dievaluasi dan hasil empiris menunjukkan bahwa jaringan syaraf tiruan adalah alat yang cukup baik untuk memprediksi

pasar keuangan.

Kata kunci: Peramalan, prediksi, indeks harga saham, indikator teknis, jaringan syaraf tiruan

pengambilan keputusan keuauangngan investor. IItutu sebabnya memprediksi gerakan yang tepat dari indeeksks harga saham jauh dianggap. Peneneliltian ini bertujuan untuk mengevaluaasisi efektivitas pengggununaaaan n inindidikakatoor r teknis, sepertii A A / D Oscillator, Movingng Average, RSRSI,I, CCCCI, MACD, dll dalal m memempmprediksi pergererakan Bursa Effeek Indekk ekss HaHarrga Inndodonesia (BEI). Sebuah jariningagan syyararafaf ttiri uan diigug nakan untuk kpeperarammalann indeks harga saham. Data yang ada dicapaiai darii YaYahohoo.oFinanance. Un

Untut kk menanangkap hubungan antara indikator teknis dan tingkaatt indeeksks di pasaar un

untutuk peperiode diselidiki, jaringan saraf propagasi kembali digunanakann.. KiKinnerja statistikk dan keuangan dari teknik ini dievaluasi dan hasil empiris mmenunjukkakann bahwaa jaringan syaraf tiruan adalah alat yang cukup baik untuk mmempreeddikssi

pasarr kkeuangganan.

Ka

Katata kkunci: Peramalan, prediksi, indeks harga saham, indikator teknknisis, jajarringngan sy

iii ABSTRACT

Stock price index is the initial significant factor influencing on investors' financial decision making. That's why predicting the exact movements of stock price index is considerably regarded. This study aims at evaluating the effectiveness of using technical indicators, such as A/D Oscillator, Moving Average, RSI, CCI, MACD, etc. in predicting movements of Indonesian Stock Exchange Price Index (IDX). An artificial neural network is employed for stock price index forecasting. The existing data are achieved from Yahoo.Finance. To capture the relationship between the technical indicators and the levels of the index in the market for the period under investigation, a back propagation neural network is used. The statistical and financial performance of this technique is evaluated and empirical results revealed that artificial neural networks are fairly good tools for financial market predicting.

Keywords: Forecasting, prediction, stock price index, technical indicators, artificial neural networks (ANN)

Stock price index is the initialalssigignificant ffacctotor rinfluencing on investors' financial decision making. TThahat's why predicting the exact movemements of stock price index is considerablbly regarded. This sstut dydy aimims s atatevavaluating the effefectciveness of using technicacal indicators, susuchch aas A/D Oscillator, MoM vivingngAAvverage, RSI,CCCI, MACD, ettcc. in prededicictitingng movemements of Indonesian Stockck EExchangnge PrPricice Indexx (IDX). An artrtifificiciaial neeuural network is employed for stock price iindndex fororececasastitng. ThT e exxisistiting datata are achieved from Yahoo.Finance. To capture the rrelelataionshiipp be

betwtween n the technical indicators and the levels of the index in the mmarkketet fforor the period under investigation, a back propagation neural network iss usedd. TThehe statistiical and financial performance of this technique is evaluated annd emppiiricaal resultltss revealed that arartitifificicialal nneural networksks areare ffaiairlrly y good tools ffoor financicialal ma

m rket predicting.

Ke

Keywywords: FForecaastinstingg, predidictctioion,n, sstotockck price iindndexe , ttechhnical ini indidicacators, artifificicialal neural l nenettworks (ANN)

iv

ACKNOWLEDGEMENTS

I would like to express my sincere thanks and appreciation to my supervisor, Professor Dr. J. Sukmawati Sukamulja, for her valuable advice, guidance and very kind support from the beginning of my study at Faculty of Master of Management until my graduation.

My gratitude to Drs. Felix Wisnu Isdaryadi, MBA for his sincere comments for the final edition of this thesis.

I would like to express myy ssininccere thanknkss ana d appreciation to my supervisor, Professor Dr. J. Sukkmmawati Sukamulja, for her valuablele aadvice, guidance and very kind supporttttffrom the beginning gofofmmy ystudstudyy att Faculty of Maststere of Management until mymy graduation.

M

My gratititutudede tto Drs.s. Felix Wisnu Isdaryadi, MBA foF forr his sincncereree commenents for the fifinanal lede itioonn of this thesis.

v Table of Contents DECLARATION ... i INTISARI ...ii ABSTRACT ...iii ACKNOWLEDGEMENTS ... iv

List of Tables ...vii

List of Figures ...viii

ABBREVATIONS ... ix

CHAPTER 1 INTRODUCTION ... 1

1.1. Problem Identification ... 5

1.2. Objective of the Research ... 6

1.4. Scope of the Research ... 8

1.5. Organization of the Thesis ... 9

CHAPTER 2 LITERATURE REVIEW ... 10

2.1 Artificial Neural Network ... 10

2.2 Review of previous researches ... 11

2.3 Learning Paradigms in ANNs ... 14

CHAPTER 3 RESEARCH METHODOLOGY ... 20

3.1 Statistical Performance Evaluation of the Model... 22

DECLARATION ... INTISARI ... ABSTRAACCT ... ACCKKNOWLELEDGD EMMENTSS.... ... Listt of f TaTablbles ... Li Liststoof fFigugures ...v AB

ABBRREEVATIONS ... CHAAPTER 1 INTRODUCTION ...

1.

11.PrP oblem Identification ... 1.2. Objective off the Researchh ... 1.

1 4.4.Scope of the Research ... 1.

1.5.5 Organizatii tion oof f ththee Thesis ... CHAPAPTETERR 22LILITERATURE REVEVIEW ... 2.1 Artificial Neural Network... 2.2 Review of previous researchehes ... 2.3 Learning Paradigms in ANNs ...

vi

3.2 Financial Performance Evaluation of the Model ... 24

3.3 Research Data... 25

3.4 Data preparation ... 26

3.5 Variable Calculation... 27

CHAPTER 4 DESCRIPTIVE STATISTICS ... 31

CHAPTER 5 RESEARCH RESULTS AND ANALYSIS ... 36

5.1 Comparison of Financial Performance... 36

5.2 Comparison of Statistical Performance ... 45

CHAPTER 6 CONCLUSION ... 49

REFERENCES ... 54

Apendix A: Matlab code... 58

A. Preprocess code ... 58

B. Training code ... 60

C. Testing code... 73 3.4 Data preparation ... 3.5 VariableeCCalculation... CHAPPTTER 4 DESSCRCRIPIPTTIVESSTATATITISTSTICCSS... CH

CHAPTEER R 55 RRESEARCRCHH RESULTS AND ANNALALYSIS... 5.1 CComomparisoson of Financial Performance... 5.

5.22 Commparison of Statistical Performance ... CH

C APPTER 6 CONCLUSION ... REFEERENCES ...

Ap

Apenendix A: Matlab code... A. Preprocess code ... B.

B TrTraiaininingng code ... C.

vii List of Tables

Table 1. The number of sample in the entire data set ... 26

Table 2. Selected technical indicators and their formulas ... 28

Table 3. Defined Variables ... 30

Table 4. ANN parameter levels tested in parameter setting ... 32

Table 5. Summary statistics for the selected indicators ... 33

Table 6. Three parameters for training and testing of ANN model ... 37

Table 7: Testing with parameter combination (10, 0.2 , 0.5, 1e6) ... 38

Table 8. Testing with parameter combination (30, 0.3, 0.5, 1e6) ... 39

Table: 9. Testing with parameter combination (50, 0.2, 0.5, 1e-6) ... 39

Table 10. Summary of the best forecasting, parameters (10, 0.2 , 0.5, 1e6) ... 41

Table 11. Financial performance of ANN model ... 42

Table 12. The empirical result of other research ... 44

Table: 13 the best statistic & financial performance ... 46

Table 14. Statistical performance of ANN model ... 48 Table 1. The number offssaample in the entire datatasset ... Table 2. Selecctetedd technical indidcators and their formulas ... Table 3.3. Defined VVarariaiablblese ... Ta

Table 4. AANNNNpparamettererllevels tested in parameteterr setting ... Tablble e 5.5 SSummmary statistics for the selected indicators ... Ta

Tablbe 6. TThree parameters for training and testing of ANN modeell... Ta

T ble 77: Testing with parameter combination (10, 0.2 , 0.5, 1e6) ... Tablee 8. Testing with parameter combination (30, 0.3, 0.5, 1e6) ...

Ta

Tablbee: 9. Testing with pparameter combination ((50,,0.2, 0.5, 1e-6) ... Table 10. Summary of the best forerecac ststining, parameters (10, 0.2 , 0.5, 1e6) ... Ta

Tablble e1111. Financial pep rformance of ANN model ... Ta

Tablblee 12. The emempipiricai all reresusultlt of ototheher r rresearrchch ... Table: 13 the best statistic & finanancial peerfr ormance ... Table 14. Statistical performancce of ANNmmodel ...

viii List of Figures

Fig. 1 An artificial neural network is an interconnected group of nodes... 11

Fig. 2 A Neural network with three-layer feed forward ... 16

Fig. 3 Tan-Sigmoid Transfer Function and Linear Transfer Function ... 31

Fig. 4 Data preparation (actual technical parameters & normalized technical parameters) ... 34

Fig. 5 Training process of ANN model ... 34

Fig. 6 Testing of ANN model ... 35

Fig.7 Predict next trading day, by entering new data to the network ... 35

Fig. 8 Training & Forecasting performance (%) of ANN model for a whole data set (n = 50, η = 0.2, μ = 0.5, ep = 1e6). ... 41

Fig. 9 Forecasting performance (%) of ANN model for various η values ... 43 Fig. 1 An artificial neurralal network is an interconnnnected group of nodes... Fig. 2 A Neurarall network withh three-layer feed forward ... Fig. 3 TTan-SigmoididTTrarannsfer FuFuncnctitionon andnd LiLinenearar TTrar nsfer Funcctition ... Fi

Fig. 4 Datataa prprepeparation n (a(actual technical parameetetersr & norrmamalilizezed techninical pa

pararammeteersrs) ...

Fi

Figg.55 Traraining process of ANN model ... Fi

Fig. 6 Testing of ANN model ...

Fig.77 Predict next trading day, by entering new data to the network ... Fi

Fig.g 88 Training & Forecasting performance (%) of ANN model for awwhoolele datta a set (n = 5050, η = 00.22, μ ==00.5.5,,ep = 11e6e6)). ... Fi

ix

ABBREVATIONS

GDP : gross domestic product

IA : artificial intelligent

ANN : artificial neural network

IDX : Indonesian Stock Index

JKSE : Jakarta Stock Exchange (Pervious name of IDX)

MAE : mean absolute error

RMSE : root mean square error

MAPE : mean absolute percentage error

R2 : goodness of fit

APE : absolute percentage error

PO : predicted output

AO : actual output

CCI : commodity channel index

MACD: moving average convergence divergence

ROC : price-rate-of change

RSI : relative strength index GDP : gross domestic prodducuctt

IA : artificiaall iintelligent

ANN :: artificial neuru alalnneetworkk

IDDX : Inndodonenesisian Stotockck Index

JKSEE :JJakarartta Stock Exchange (Pervious name of IDX)

MA

MAE E :mmean absolute error

RMSEE : root mean square error

MA

MAPEE : mean absolute percentage error

R2 : goodness of fit

AP

APEE : ababsosolulutetepperercecentntagagee ererror r

POO ::ppreredidictcteded ooututpput

AO : actual output

CCI : commodity channel indexx

x PR : predicted rate (forecasting rate)

n : neuron

η : learning rate μ : momentum constant

ep : epoch

IT : information technology

LSM : The Libyan Exchange Stock Market TEPIX : The Tehran Exchange Price Index η : learning rate

μ : momentum coconstant ep : epochh

IT : iinformationnttecechnhnology y

LSMM : The LiLibybyan Exchangngee StStockk Markkett T