Title: Forecasting the BET-C Stock Index with Artificial Neural Networks
1Forecasting the BET-C Stock Index with Artificial
Neural Networks
DOCTORAL SCHOOL OF FINANCE AND BANKING
DOFIN ACADEMY OF ECONOMIC STUDIES
MSc Student Stoica Ioan-Andrei Supervisor
Professor Moisa Altar
July 2006
2Stock Markets and Prediction
- Predicting stock prices - goal of every investor
trying to achieve profit on the stock market - predictability of the market - issue that has
been discussed by a lot of researchers and
academics - Efficient Market Hypothesis - Eugene Fama
- three forms
- Weak future stock prices cant be predicted
using past stock prices - Semi-strong even published information cant be
used to predict future prices - Strong market cant be predicted no matter what
information is available
3Stock Markets and Prediction
- Technical Analysis
- castles-in-the air
- investors behavior and reactions according to
these anticipations - Fundamental Analysis
- firm foundations
- stocks have an intrinsic value determined by
present conditions and future prospects of the
company - Traditional Time Series Analysis
- uses historic data attempting to approximate
future values of a time series as a linear
combination - Machine Learning - Artificial Neural Networks
4The Artificial Neural Network
- computational technique that benefits from
techniques similar to those employed in the human
brain - 1943 - W.S. McCulloch and W. Pitts attempted to
mimic the ability of the human brain to process
data and information and comprehend patterns and
dependencies - The human brain - a complex, nonlinear and
parallel computer - The neurons
- elementary information processing units
- building blocks of a neural network
5The Artificial Neural Network
- semi-parametric approximation method
- Advantages
- ability to detect nonlinear dependencies
- parsimonious compared to polynomial expansions
- generalization ability and robustness
- no assumptions of the model have to be made
- flexibility
- Disadvantages
- has the black box property
- training requires an experienced user
- training takes a lot of time, fast computer
needed - overtraining ? overfitting
- undertraining ? underfitting
6The Artificial Neural Network
7The Artificial Neural Network
8The Artificial Neural Network
9The Artificial Neural Network
Undertraining/Underfitting
10Architecture of the Neural Network
- Types of layers
- input layer number of neurons number of inputs
- output layer number of neurons number of
outputs - hidden layer(s) number of neurons trial and
error
- Connections between neurons
- fully connected
- partially connected
- The activation function
- threshold function
- piecewise linear function
- sigmoid functions
11The feed forward network
m number of hidden layer neurons n number of
inputs
12The Feed forward Network with Jump Connections
13The Recurrent Neural Network - Elman
allows the neurons to depend on their own lagged
values ? building memory in their evolution
14Training the Neural Network
Objective minimizing the discrepancy between
real data and the output of the network
O - the set of parameters ? loss function
? nonlinear ? nonlinear optimization problem
- backpropagation
- genetic algorithm
15The Backpropagation Algorithm
- alternative to quasi-Newton gradient descent
- O0 randomly generated
- ? learning parameter, in .05,.5
- after n iterations µ0.9, momentum parameter
- problem local minimum points
16The Genetic Algorithm
- based on Darwinian laws
- Population Creation N random vectors of weights
- Selection ? (Oi Oj) parent vectors
- Crossover Mutation ? C1,C2 children vectors
- Election Tournament the fittest 2 vectors passed
to the next generation - Convergence G generations
- G - large enough so there are no significant
changes in the fitness of the best individual for
several generations
17Experiments and Results
Data
- BET-C stock index daily closing prices, 16
April 1998 until 18 May 2006 - daily returns
- conditional volatility - rolling 20-day standard
deviation - BDS-Test for nonlinear dependencies
- H0 i.i.d. data
- BDSm,eN(0,1)
Series m2 m2 m3 m3 m4 m4
e1 e1.5 e1 e1.5 e1 e1.5
OD 16.6526 17.6970 18.5436 18.7202 19.7849 19.0588
ARF 16.2626 17.2148 18.3803 18.4839 19.7618 18.9595
18Experiments and Results
- 3 types of Ann's
- feed-forward network
- feed-forward network with jump connections
- recurrent network
- Input Rt-1 Rt-2 Rt-3 Rt-4 Rt-5 Vt
- Output next-day-return Rt
- Training genetic algorithm backpropagation
- Data divided in
- training set 90
- test set 10
- one-day-ahead forecasts - static forecasting
- Network
- trained 100 times
- best 10 SSE
- best 1 - RMSE
19Experiments and Results
Evaluation Criteria
- In-sample Criteria
- Out-of-sample Criteria
- Pesaran-Timmerman Test for Directional Accuracy
- H0 signs of the forecast and those of the real
data are independent - DAN(0,1)
20Experiments and Results
- ROI - trading strategy based on the sign
forecasts - buy sign
- - sell sign
- Finite differences
Benchmarks
- Naïve model Rt1Rt
- buy-and-hold strategy
- AR(1) model LS overfitting
- RMSE
- MAE
21Experiments and Results
Naïve AR(1) FFN no vol FFN FFN-jump RN
R2 - 0.079257 0.083252 0.083755 0.084827 0.091762
SSE - 0.332702 0.331258 0.331077 0.330689 0.328183
RMSE 0.015100 0.011344 0.011325 0.011304 0.011332 0.011319
MAE 0.011948 0.008932 0.008929 0.008873 0.008867 0.008892
HR 55.77 (111) 56.78 (113) 57.79 (115) 59.79 (119) 59.79 (119) 59.79 (119)
ROI 0.265271 0.255605 0.318374 0.351890 0.331464 0.412183
RP 15.02 14.47 18.02 19.92 18.77 23.34
PT-Test - - 14.79 15.01 15.01 14.49
BH 0.2753 0.2753 0.2753 0.2753 0.2753 0.2753
FFN FFN-jump RN
Volatility -0.1123 -0.1358 -0.1841
22Experiments and Results
Actual, fitted ( training sample)
23Experiments and Results
Actual, fitted ( test sample)
24Conclusions
- RMSE and MAE lt AR(1) ? no signs of overfitting
- R2 lt 0.1 ? forecasting magnitude is a failure
- sign forecasting 60 ? success
- Volatility
- improves sign forecast
- finite differences ? negative correlation
- perceived as measure of risk
- trading strategy outperforms naïve model and
buy-and-hold - quality of the sign forecast confirmed by
Pesaran-Timmerman test
25Further development
- Volatility other estimates
- neural classificator specialized in sign
forecasting - using data outside the Bucharest Stock Exchange
- T-Bond yields
- exchange rates
- indexes from foreign capital markets