Tzu-Wen Kuo

Evolutionary Computation in Economics and Finance: A Bibliography

This chapter presents a bibliography on the application of evolutionary computation to economics and finance. Publications included in this bibliography are classified by application domain, published journal or conference proceedings. Information on some useful websites and software is also provided.

Discovering Financial Technical Trading Rules Using Genetic Programming with Lambda Abstraction

We applied genetic programming with a lambda abstraction module mechanism to learn technical trading rules based on SP each rule uses a combination of one to four widely used technical indicators to make trading decisions. The consensus among these trading rules is high. For the majority of the testing period, 80% of the trading rules give the same decision. These rules also give high transaction frequency. Regardless of the stock market climate, they are able to identify opportunities to make profitable trades and out-perform buy-and-hold.

Genetic Programming: A Tutorial With The Software Simple GP

This chapter demonstrates a computer program for tutoring genetic programming (GP). The software, called Simple GP, is developed by the AI-ECON Research Center at National Chengchi University, Taiwan. Using this software, the instructor can help students without programming background to quickly grasp some essential elements of GP. Along with the demonstration of the software is a list of key issues regarding the effective design of the implementation of GP. Some of the issues are already well noticed and studied by financial users of GP, but some are not. While many of the issues do not have a clear-cut answer, the attached software can help beginners to tackle those issues on their own. Once they have a general grasp of how to implement GP effectively, many advanced materials prepared in this volume are there for further exploration.

Overfitting or poor learning: a critique of current financial applications of GP

Motivated by a measure of predictability, this paper uses the extracted signal ratio as a measure of the degree of overfitting. With this measure, we examine the performance of one type of overfitting-avoidance design frequently used in financial applications of GP. Based on the simulation results run with the software Simple GP, we find that this design is not effective in avoiding overfitting. Furthermore, within the range of search intensity typically considered by these applications, we find that underfitting, instead of overfitting, is the more prevalent problem. This problem becomes more serious when the data is generated by a process that has a high degree of algorithmic complexity. This paper, therefore, casts doubt on the conclusions made by those early applications regarding the poor performance of GP, and recommends that changes be made to ensure progress.

A GENETIC PROGRAMMING APPROACH TO MODEL INTERNATIONAL SHORT-TERM CAPITAL FLOW

We model international short-term capital flow by identifying technical trading rules in short-term capital markets using Genetic Programming (GP). The simulation results suggest that the international short-term markets was quite efficient during the period of 1997–2002, with most GP generated trading strategies recommending buy-and-hold on one or two assets. The out-of-sample performance of GP trading strategies varies from year to year. However, many of the strategies are able to forecast Taiwan stock market down time and avoid making futile investment. Investigation of Automatically Defined Functions shows that they do not give advantages or disadvantages to the GP results.

Are Efficient Markets Really Efficient?: Can Financial Econometric Tests Convince Machine-Learning People?

Using Quinlan’s Cubist, this paper examines whether there is a consistent, interpretation of the efficient market hypothesis between financial econometrics and machine learning. In particular, we ask whether machine learning can be useful only in the case when the market is not efficient. Based on the forecasting performance of Cubist in our artificial returns, some evidences seems to support this consistent interpretation. However, there are a few cases whereby Cubist can beat the random walk even though the series is independent. As a result, we do not consider that the evidence is strong enough to convince one to give up his reliance on machine learning even though the efficient market hypothesis is sustained.