![Page 1: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/1.jpg)
NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI
授課教師:楊婉秀報告人:李宗霖
![Page 2: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/2.jpg)
Outline• Introduction
• Forecasting the Stock Market
• Neural Network and its Usage in the Stock Market
• A Case Study on the Forecasting of the KLCI
• Discussion
• Conclusion and Future Research
2
![Page 3: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/3.jpg)
Introduction
3
![Page 4: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/4.jpg)
Introduction• It is generally very difficult to forecast the movements of
stock markets.
• Artificial neural network is a well-tested method for financial analysis on the stock market.
4
![Page 5: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/5.jpg)
Introduction• Using neural networks in equity market applications
include:
• Forecasting the value of a stock index
• Recognition of patterns in trading charts
• Rating of corporate bonds
• Estimation of the market price of options
• Indication of trading signals of selling and buying
5
![Page 6: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/6.jpg)
Introduction• Feed-forward back propagation
• Without the use of extensive market data or knowledge useful prediction can be made and significant paper profit can be achieved.
6
![Page 7: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/7.jpg)
Forecasting the Stock Market
7
![Page 8: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/8.jpg)
Forecasting the Stock Market• There are three schools of thought in terms of the ability
to profit from the equity market.
• Random Walk Hypothesis & Efficient Market Hypothesis
• fundamental analysis
• Technical analysis
8
![Page 9: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/9.jpg)
Forecasting the Stock Market• Use a variety of techniques to obtain multiple signals.
• Neural networks are often trained by both technical and fundamental indicators to produce trading signals.
9
![Page 10: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/10.jpg)
Forecasting the Stock Market• For fundamental methods:
• retail sales
• gold prices
• industrial production indices
• foreign currency exchange rates
• For technical methods:
• delayed time series data
• technical indicators
10
![Page 11: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/11.jpg)
Neural Network and its Usage in the Stock Market
11
![Page 12: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/12.jpg)
3.1 Neural networks• Neural networks
12
![Page 13: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/13.jpg)
3.2 Time series forecasting with neural networks
13
![Page 14: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/14.jpg)
3.2 Time series forecasting with neural networks• Three major steps in the neural network based forecasting
proposed in this research:
• Preprocessing
• Architecture
• Postprocessing
14
![Page 15: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/15.jpg)
3.3 Measurements of neural network training
• Normalized Mean Squared Error (NMSE)
• Signs
• Gradients
15
![Page 16: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/16.jpg)
• We argue that NMSE may not be the case for trading in the context of time series analysis.
3.3 Measurements of neural network training
16
![Page 17: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/17.jpg)
3.3 Measurements of neural network training
17
![Page 18: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/18.jpg)
A Case Study on the Forecasting of the KLCI
18
![Page 19: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/19.jpg)
A Case Study on the Forecasting of the KLCI• Kuala Lumpur Composite Index (KLCI)
• Neural networks are trained to approximate the market values.
• To find the hidden relationship between technical indicators and future KLCI.
19
![Page 20: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/20.jpg)
Data choice and pre-processing
20
![Page 21: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/21.jpg)
Data choice and pre-processing• The major types of indicators
• moving average (MA)
• momentum (M)
• Relative Strength Index (RSI)
• stochastics (%K)
• moving average of stochastics (%D)
21
![Page 22: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/22.jpg)
Data choice and pre-processing
• inputs• It−1
• It
• MA5
• MA10
• MA50
• RSI• M• %K• %D
• output• It+1
22
![Page 23: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/23.jpg)
Data choice and pre-processing• Normalization
23
![Page 24: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/24.jpg)
Nonlinear analysis of the KLCI data
24
![Page 25: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/25.jpg)
Nonlinear analysis of the KLCI data
25
![Page 26: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/26.jpg)
Nonlinear analysis of the KLCI data
26
![Page 27: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/27.jpg)
Neural network model building• Historical data are divided into three parts
• Training sets (2/3)
• Validation sets (2/15)
• Testing sets (3/15)
27
![Page 28: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/28.jpg)
Neural network model building• A trade-off between convergence and generalization.
• Number of hidden nodes
28
![Page 29: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/29.jpg)
Neural network model building
29
![Page 30: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/30.jpg)
Neural network model building• Primary sensitive analysis is conducted for input
variables.
• Low influence factors: M20, M50, %K, %D
• High influence factors: It, RSI, M, MA5
• Chosen for input: It, MA5, MA10, RSI, M, It−1
30
![Page 31: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/31.jpg)
Neural network model building
31
![Page 32: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/32.jpg)
Neural network model building
32
![Page 33: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/33.jpg)
Paper profits using neural network predictions
33
![Page 34: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/34.jpg)
Paper profits using neural network predictions
34
![Page 35: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/35.jpg)
Paper profits using neural network predictions• In a real situation, this might not be possible as some
indexed stocks may not be traded at all on some days.
• The transaction cost of a big fund trading, which will affect the market prices was not taken into consideration in the calculation of the “paper profit”.
35
![Page 36: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/36.jpg)
Benchmark return comparison• Benchmark 1:
• Benchmark 2:
36
![Page 37: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/37.jpg)
Benchmark return comparison• Benchmark 3:
37
![Page 38: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/38.jpg)
Comparison with ARIMA• Autoregressive Integrated Moving Average (ARIMA)
Model
• Introduced by George Box and Gwilym Jenkins in 1976.
• Provided a systematic procedure for the analysis of time series that was sufficiently general to handle virtually all empirically observed time series data patterns.
• ARIMA(p, d, q)
38
![Page 39: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/39.jpg)
39
![Page 40: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/40.jpg)
Discussion
40
![Page 41: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/41.jpg)
Discussion• A very small NMSE does not necessarily imply good
generalization.
• Better testing results are demonstrated in the period near the end of the training sets.
• We have no data to test the “best” model.
• There are two approaches for using the forecasting result.
• best-so-far approach
• committee approach
41
![Page 42: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/42.jpg)
Discussion• Measures
• NMSE
• Sign
• Gradient
• Four challenges
• inputs and outputs
• types of neural networks and the activation functions
• neural network architecture
• evaluate the quality of trained neural networks for forecasting
42
![Page 43: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/43.jpg)
Conclusion and Future Research
43
![Page 44: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/44.jpg)
Conclusion and Future Research• It shows that useful prediction could be made for KLCI
without the use of extensive market data or knowledge.
• It shows how a 26% annual return could be achieved by using the proposed model.
• It highlights the following problems associated with neural network based time series forecasting:
• the hit rate is a function of the time frame chosen for the testing sets;
• generalizability of the model over time to other period is weak;
• there should be some recency trade offs.
44
![Page 45: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/45.jpg)
Conclusion and Future Research• A mixture of technical and fundamental factors as inputs
over different time periods should be considered.
• Sensitivity analysis should be conducted which can provide pointers to the refinement of neural network models.
• The characteristics of emerging markets such as KLCI should be further researched to facilitate better modeling of the market using neural networks.
45
![Page 46: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/46.jpg)
Thanks for Your Listening
46
![Page 47: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/47.jpg)
Forecasting the Stock Market• The research done here would be considered a violation
of the above two hypotheses above for short-term trading advantages in, Kuala Lumpur Stock Exchange (KLSE for short hereafter)
47
![Page 48: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/48.jpg)
Forecasting the Stock Market• There are three schools of thought in terms of the ability
to profit from the equity market.
• The first school believes that no investor can achieve above average trading advantages based on the historical and present information.
• Random Walk Hypothesis
• Efficient Market Hypothesis
• Taylor provides compelling evidence to reject the random walk hypothesis and thus offers encouragement for research into better market prediction.
48
![Page 49: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/49.jpg)
Forecasting the Stock Market• The second school's view is the so-called fundamental
analysis.
• It looks in depth at the financial conditions and operating results of a specific company and the underlying behavior of its common stock.
• In 1995 US$1.2 trillion of foreign exchange swapped hands on a typical day [10]. The number is roughly 50 times the value of the world trade in goods and services which should be the real fundamental factor.
49
![Page 50: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/50.jpg)
Forecasting the Stock Market• Technical analysis belongs to the third school of thought.
• It attempts to use past stock price and volume information to predict future price movements.
• These five series are open price, close price, highest price, lowest price and trading volume.
• Most of the techniques used by technical analysts have not been shown to be statistically valid and many lack a rational explanation for their use
50
![Page 51: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/51.jpg)
• Normalized Mean Squared Error (NMSE)
• Signs
• Gradients
3.3 Measurements of neural network training
51
![Page 52: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/52.jpg)
3.3 Measurements of neural network training
• Normalized Mean Squared Error (NMSE)
52
![Page 53: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/53.jpg)
• Signs
3.3 Measurements of neural network training
53
![Page 54: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/54.jpg)
• Gradients
3.3 Measurements of neural network training
54
![Page 55: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/55.jpg)
Data choice and pre-processing
55
![Page 56: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/56.jpg)
Data choice and pre-processing
56
![Page 57: NEURAL NETWORKS FOR TECHNICAL ANALYSIS: A STUDY ON KLCI 授課教師:楊婉秀 報告人:李宗霖](https://reader034.vdocuments.pub/reader034/viewer/2022052205/56649dac5503460f94a9c61e/html5/thumbnails/57.jpg)
Data choice and pre-processing
57