<![CDATA[The application of Artificial Intelligence (AI) with the Recurrent Neural Network (RNN) Long Short-Term Memory (LSTM) algorithm which has excellent accuracy in predicting stock prices, but still needs to be developed optimally in the accounting and finance fields. In many cases it was found that the value of the results of the traditional Capital Asset Pricing Model (CAPM) calculations was consistently below the estimated return. Therefore, it is necessary to calculate the CAPM that can optimize the estimated returns more accurately, by combining it using AI technology with the LSTM method RNN algorithm. The purpose of this study is to prove the accuracy of the CAPM calculation results generated by AI compared to the traditional CAPM calculation method in providing an estimate of the return on the best blue-chip group in the LQ45 index during the 2015-2019 observation period. The sample of this study uses data on closing prices adjusted for companies with the lowest Debt to Equity Ratio (DER). The results of the adjusted closing price prediction using AI with the LSTM method have high accuracy. From the analysis of the different paired sample t-test, it was found that there was a significant difference in the mean Absolute Percentage Error (MAPE), where the AI-optimized CAPM model had a lower MAPE value than the traditional CAPM model. This study concludes that the AI-optimized CAPM calculation method has proven to be able to provide more accurate return estimates than traditional CAPM calculation methods.]]>
Copyrights © 2022
