Using least square support vector regression with genetic algorithm to forecast beta systematic risk

作者：Yuan Fong Ching^*; Lee Chao Hui

来源：Journal of Computational Science, 2015, 11: 26-33.

DOI：10.1016/j.jocs.2015.08.004

• Beta forecasting
• Capital asset pricing model
• Least squares support vector regression
• Genetic algorithm
• Artificial neural network

Since its development, the capital asset pricing model (CAPM) has been extended into all areas of modern corporate finance and investments. The beta systematic risk, which gives the CAPM its power, is the most important factor in measuring and judging risks. The available evidence on the stability of beta indicates that the beta is generally not stable. Hence we face the problem of forecasting future betas in order to use the CAPM. If the portfolio manager cannot predict future beta coefficients, the applicability of this phase of modern capital-market theory is somewhat restricted.
This paper discusses different methods of predicting beta, using the financial statements of 88 Taiwan corporations from the period 2001 to 2010. The experimental results indicate that the most precise forecasts are given by the least square regression algorithm optimized by genetic algorithm.

• 出版日期2015-11