Professional Documents
Culture Documents
Busines Intelligence & Analytics: Artificial Neural Networks (ANN) For Financial Time Series Modeling
Busines Intelligence & Analytics: Artificial Neural Networks (ANN) For Financial Time Series Modeling
(MS6840)
Compare
Feed-Forward Neural N/w topologies
Financial Time Series Modeling:
The case of stock markets
Econometric data
} In econometrics there are three main types of data (not
necessarily mutually exclusive)
} Cross-sectional data
} Time series data
} Panel (longitudinal) data
9000
8000
7000
6000
5000
4000
3000
2000
1000
0 12 24 36 48 60 72 84 96 108 120 132 144 156 168 180 192 204 216 228 240 252 264 276 288 300 312 324 336 348 360 372 384 396400
Time
Q1 Q3 Q1 Q3 Q1 Q3
Trend
Seasonality
Y Y
Time time
Cycle Noise
Time series classical models
} Multiplicative
} Ûi=Ti*Si*Ci*Ii Ti: Trend component
Si: Seasonal component
} Additive Ci: Cyclical component
Ii: Irregular component
} Ûi=Ti+Si+Ci+Ii
} Trend
} Û =bXi+c
i
} Seasonality
} Ûi=b1Xi + b2Q1+ b3Q2 + b3Q3+c
} Seasonality index
Forecasting stock market time series
} Efficient Market Hypothesis
} Week
} Semi-strong
} Strong
} Random Walk Hypothesis
} Implication: Only luck???
Stock Price Forecasting
} Fundamental Analysis
} External factors
} Gold price, exchange rate, oil price etc.
} Technical Analysis
} Moving averages
} Relative strength indicator
} Momentum, M= CCP-OCP
} Stochastic Oscillator
} Price rate of Change Indicator
ANN modeling process
} Data preparation
} Network selection
} Training
} Testing and validation
} Apply and re-train
Sources of Data
} Financial times
} Bloomberg
} Wall street journal
} Reuters
} Yahoo! finance
Data preparation
Training Testing validation
} Input, output selection
} Data Transformation
} Range (say [-1,1])
} First differences
} Logarithmic
Compare
Model performance & prediction errors
} Mean Square Error (MSE)
} Root Mean Square Error (RMSE)
} Mean Absolute Error (MAE)
} Mean Absolute Percentage Error (MAPE)