Download presentation
Presentation is loading. Please wait.
1
Non-fixed and Asymmetrical Margin Approach to Stock Market Prediction using Support Vector Regression Haiqin Yang, Irwin King and Laiwan Chan Department of Computer Science and Engineering The Chinese University of Hong Kong November 18-22, 2002 ICONIP ’ 02
2
2/25 Index Motivation SVR Introduction Approach Conclusion Experiments & Results
3
3/25 Motivation Combine them: Non-fixed and Asymmetrical margin Two characteristics: fixed and symmetrical Predictive accuracy only? Downside risk!
4
4/25 Support Vector Regreesion (SVR) introduction Developed by Vapnik (1995) Developed by Vapnik (1995) Model: Model: estimate objective function: minimize train data:
5
5/25 SVR Introduction (Cont ’ d) Loss function: The objective function f is represented by the dotted points.
6
6/25 Related Applications Support Vector Method for Function Approximation, Regression Estimation and Signal Processing (Vapnik et al., 1996) Support Vector Method for Function Approximation, Regression Estimation and Signal Processing (Vapnik et al., 1996) Predicting time series with support vector machine (Muller et al., 1997) Predicting time series with support vector machine (Muller et al., 1997) Application of support vector machines to financial time series forecasting (E.H.Tay and L.J.Cao. 2001) Application of support vector machines to financial time series forecasting (E.H.Tay and L.J.Cao. 2001)
7
7/25 Approach Two characteristics: 4 kinds of margins Symmetrical Asymmetrical Fixed Non-fixed fixed,symmetrical. FASM NASM FAAM NAAM + + + + + +
8
8/25 Previous setting Previous others’ method Symmetrical Asymmetrical Fixed Non-fixed Symmetrical Asymmetrical Fixed FASMFAAM Non-fixedNASMNAAM In our previous work: Support Vector Machines Regression for volatile stock market prediction (IDEAL’02)
9
9/25 New Approach Two characteristics of the margin in – insensitive loss function: fixed and symmetrical. Non-fixed Asymmetrical Symmetrical Fixed
10
10/25 Formulas A general type of –Insensitive loss function Fixed and Symmetrical Margin (FASM): Fixed and Asymmetrical Margin (FAAM): Non-fixed and Symmetrical Margin (NASM): Non-fixed and Asymmetrical Margin (NAAM): up margin down margin
11
11/25 Formulas QP problem: s.t. Objective function: Kernel function: e.g. RBF
12
12/25 How to set margin? Margin width: Up margin: Down margin:
13
13/25 Experiment Accuracy Metrics MAE: UMAE: DMAE: actual value, predictive value number of testing data Total error Upside risk Downside risk
14
14/25 Experiment Description Model: Data: Hang Seng Index (HSI), Dow Jones Industrial Average (DJIA). Time periods: Jan. 2, 1998 ~ Dec. 29, 2000 (3 years) Ratio of training data and testing data: 5:1. Procedures: one day ahead prediction. Environments CPU: Pentium 4, 1.4 G Memory: RAM 512M OS: Windows2000 Time: few hours.
15
15/25 Experiment Description Three kinds of experiments Test the effect of parameters in NAAM to obtain a better result. Compare the result of NAAM with NASM, AR(4), RBF network (also test the effect of the number of hidden units). Compare the results of NAAM, NASM with FASM and FAAM.
16
16/25 Actual Parameter Setting
17
17/25 Effect of Length of EMA in NAAM HSI Error DJIA Error
18
18/25 Graphes HSI DJIA Data Setratio HSI100182.2820.800.114 DJIA 3079.9515.640.196
19
19/25 Effect of in NAAM HSI Error DJIA Error
20
20/25 k Effect of k in NAAM HSI Error DJIA Error
21
21/25 Comparison Results HSI Error
22
22/25 Results DJIA Error
23
23/25 NAAM, NASM vs. FASM, FAAM Fixed Margin: HSI Error Step:
24
24/25 NAAM, NASM vs. FASM, FAAM Fixed Margin: DJIA Error Step:
25
25/25 Conclusion Propose non-fixed and asymmetrical margin (NAAM) approach in SVR to predict stock market. Compare this method to non-fixed symmetrical margin (NASM) approach, AR(4), RBF network. NAAM, NASM outperform AR(4), RBF network. NAAM can reduce the downside risk. NAAM, NASM outperform FASM, FAAM.
Similar presentations
© 2024 SlidePlayer.com. Inc.
All rights reserved.