Empirical Study of Stock Selection Model in America Stock Market

• Main Authors: Yen, Yi-Chun, 顏依君
• Other Authors: Chang, Wen-Chih
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/96551316876785775230

碩士 === 中華大學 === 資訊管理學系碩士班 === 100 === This paper aims to study the performance of stock in America market of single and multi-factor stock selection models, and these models under various simulation parameters, samples based on size, as well as during different period, months, and bull/bear market. The test samples consist of all the listed stocks in America stock market. The back test period is from the third quarter of 1990 to the third quarter of 2010, a total of 20 years. The results showed that (1) The greater the ROE and GVI, the higher the return of portfolio. The smaller the P/B, P/E, P/S, the higher the return of portfolio. (2) Combining several factors with the scoring approach can raise the return of the long portfolio, and can reduce the return of the short portfolio. Especially, the combination of the ROE factor and the P/B factor. (3) The returns of equal weighted portfolio are higher than those of market value weighted portfolio. (4) The returns of the monthly trading strategy are often higher than those of quarterly trading strategy. (5) When reducing the picking percentage from 10% to 1%, the return mostly do not change, even get lower, not as expected. However, the P/S and ROE are exception. The lower the picking percentage, that is, the more extreme, the higher the rate of return. (6) The returns of large-cap stocks are often lower than those of small-cap stocks. (7) When dividing the data set into four five-year period data set, the best factor in every data set is different. However, the P/B representing value perform better than the ROE representative growth in three five-year periods except the 1996-2000 period. (8) Overall, the month effect exists, but no significant January effect exists. The performance of all stock selection factors in the first six months is better than in the second six months every year. The P/B representing value property perform poor in the months when ROE representing growth property perform well; and vice versa. (9) The stock selection effect of value stocks is rather obvious in bear market, while the stock selection effect of growth stocks is rather obvious in bull market. (10) The “buy long in bull market and sell short in bear market” strategy and the “buy long in Bull market” strategy can not beat the simple “buy long” strategy. But the “buy long in Bull market” strategy can reduce the standard deviation of the monthly return rate.