Comparison of CAPM and APT pricing models in the Kenyan stock market: the Kenyan banking sector

Abstract

This study aims to comparatively look at the CAPM and APT models and assess which one has a more explanatory power in predicting stock returns of listed banks in the NSE. This study uses secondary data from 2016 to 31st June 2024. For CAPM, regressions were done directly with the market risk being the NSE – All Share Index. For the APT model, several factors such as bank regulations and central bank policies, consumer behavior and monetary policies, were assumed to have an impact on the stock price movement. Regressions on the same were conducted as well as tests for heteroskedasticity, multicollinearity, autocorrelation and stationarity. The MAD of the two models were conducted as well and it was established that the APT model had a lower MAD value than CAPM. This implies that monetary policy makers, researchers, regulators and financial market participants can rely on the changes in the policies and regulations to assess stock price movements.