Comparative Analysis of The Accuracy of The Capital Asset Pricing Model (CAPM) and Arbitrage Pricing Theory (APT) In Predicting Stock Returns In Bank Persero (BUMN) In Indonesia

Abstract

This study aims to predict stock returns. This research takes time series data from 2018 to 2022. The sample used throughout Bank Limited (SOE). This study uses a VAR approach by comparing the CAPM with APT. The results showed that the estimated using Autoregression Vector (VAR), variables other than the variable itself most have the greatest contribution to the BETA is EXCHANGE. The variable that contributed most to the INF is RS. The variables that contributed most to the JUB JUB other than itself is RF. The variable that has the largest contribution towards EXCHANGE RATE besides itself was JUB. The variables that contributed most to the RF is EXCHANGE. The variable that contributed most to the RS is JUB. Impulse response function test results are known that the stability of all the variables that are in the period to 20 or the medium term, which creates a sense

that although there is a variable that does not affect the short-term but in the medium and long term it will affect each other. Variance Decomposition test results in the short term method of CAPM (Capital Asset Pricing Model) via RF (Risk-Free Asset Return) are more accurate in predicting stock returns of the method APT (Arbitrage Pricing Theory). In medium terms the method CAPM (Capital Asset Pricing Model) via RF (Risk-Free Asset Return) is more accurate in predicting stock returns of the method APT (Arbitrage Pricing Theory). In the long term method of CAPM (Capital Asset Pricing Model) also via RF (Risk-Free Asset Return) is more accurate in predicting stock returns of the method APT (Arbitrage Pricing Theory). Specification models created using the Roots of Characteristic Polynomial and Inverse Characteristic Roots of AR polynomial obtained stable results, it can be shown that all units are in the loop picture roots Roots Inverse Characteristic of AR polynomial.