Analysis of Size Portfolios and Risk Factor Loadings of Asset Pricing Models – Threshold Regression Approach

Abstract

The objective of the study is to analysis portfolio returns and the performance of asset pricing models in the context of macroeconomic variables. Portfolios have been constructed on the basis of firms’ size; two equally weighted decile portfolios and two value weighed decile portfolios representing the cluster of the largest and smallest firms are included in the analysis along with the market portfolio. Portfolio returns are regressed with the factor loadings of CAPM, three-factor and five-factor Fama and French asset pricing models. The relationship of portfolio returns and asset pricing models is evaluated in the presence of four macroeconomic variables namely interest rate, industrial production, producer price index and exchange rate; each macroeconomic variable is introduced separately in threshold regression to the identify regime shift effect. Analyses are performed on monthly returns of securities listed on PSX during the period from 2000 to 2010. Results of the study have revealed that only the interest rate and exchange rate are found to have a threshold effect on the portfolio returns. According to the results, the threshold effect is frequently captured through three factors Fama French model. Finally, the results also suggest that the threshold effect is only evident with large firms’ portfolios.