AN APPLICATION OF THE ARBITRAGE PRICING THEORY ON SINGAPORE STOCK RETURNS : A FACTOR ANALYSIS APPROACH

Abstract

Asset pricing has generated a great deal of interest in the field of finance since the idea was first proposed. The most researched and extensively used model to date may be the single-factor, Capital Asset Pricing Model (CAPM). In spite of this, its validity has come under continuous criticisms. The multi-factor Arbitrage Pricing Theory (APT) proposed by Ross (1976) is perhaps a timely answer to these criticisms. Lately, the APT has become the focus of many researchers and scholars, and major empirical studies done in developed capital markets such as the UK and the US stock markets have generally provided evidence that the APT is a superior alternative to the CAPM in the pricing of assets. However, empirical studies of the APT in a less developed market such as the Singapore stock market is still in the developing stages. This study attempts to provide preliminary evidence to validate the APT on the estimation of Singapore stock returns. It is proposed that there are more than one economy wide factors besides the market return factor, that influence stock returns in the Singapore context. It is also proposed that the best measure for the market return factor is the Business Times Composite Index (SHARE). By selecting a sample of 43 actively traded stocks listed on the SES over the period of March 1988 to March 1993, Factor Analysis (FA) was performed on the sample stock returns to determine the number of factors that are priced. The factors are then correlated with the STII, DBS50 index-and the SHARE index to find the best estimate for the first factor, the market return factor. To justify the number of factors chosen, FA is also performed on the residuals of the of the stock returns after correcting for the market return factor (SHARE). The methodology employed weekly returns for both the sample stocks and the market indices. The time frame of 5 years covered was also segregated firstly, into two equal halves and secondly, into four equal quarters. This was done to take into account any variations in the stock market and the economy for the time frame covered, so that the findings of the study can be interpreted úwith more confidence. The findings of the Factor Analysis on the stock returns show that on the whole, four factors are significant in influencing the stock returns or in a sense priced in the market. The first factor proposed as the market return factor, contributed most to the expansion of the stock returns. However, it is noted that the other three factors also have significant explanatory powers to the stock returns. These results are based on the factor loadings of the respective factors when the Factor Analysis is performed. The results of the Factor Analysis on the residuals also yield consistent findings that at most, three other factors are priced. To justify the use SHARE index as the proxy for the market return factor, correlations between the factors and the various proposed indices such as the DBS50 index, STII and SHARE were obtained. The results show that the SHARE index returns have the highest correlation with first factor, the market return factor. The conclusion of this study suggests that there is some evidence to prove the validity of the APT for the Singapore stock returns and that the market return factor is best represented by the SHARE index. However, further empirical research and tests have to be done before any concrete conclusions can be made about the validity of the APT in the Singapore stock market.