| Stock price volatility from time to time is a problem for investor, so the prediction of stock return is crucial in investment decision. Investor needs return assessment methods to know stock that undervalued or overvalued. This study aimed to analyze the two methods prediction of stock returns that are Capital Asset Pricing Model and Arbitrage Pricing Theory. The accuracy of these two methods is measured by Mean Absolute Deviation (MAD), while Independent Sample T-Test is used to test the significance of differences accuracy between the two methods. The sample used in this study are the stocks that are consistently listed in LQ45 period 2010-2014. Stock price data and other supporting data that used are the closing price on a monthly basis. The research results in both method of CAPM and APT shows that ADRO, ITMG, BDMN, and PTBA are undervalued stocks that worth to buy, while other stocks are overvalued. CAPM method is more accurate than the APT method for predicting stock returns, because MAD CAPM value (0.46%) is smaller than MAD APT value (1.15%). The result of Independent Sample T-Test generates that there is a significant difference in accuracy between the two. |