Up to now, there are 700 stocks listed on the Indonesia Stock Exchange (IDX). An example of an index listed on the IDX is the LQ45 stock index. LQ45 index contains 45 stocks, and these stocks are grouped because of their high liquidity, large market capitalization, and good company fundamentals. Since this index is a fundamental reference index for passive investors, it is necessary to model the prediction of LQ45 stocks to predict these stocks' movement accurately. One form of modeling used is the ARIMA (p, d, q) model. In this study, ARIMA (p, d, q) modeling was conducted to predict the price of 45 stocks in LQ45. The ARIMA (p, d, q) model is a time series model that is suitable for modeling LQ45 stocks because it is only based on the relationship with previous data (AR), and the resulting error is assumed to be related to the previous error (MA). The problem that arises is that the movement of shares can be seen from the visual image (plot), while the function of the movement is unknown. This modeling is expected to help observe the ARIMA functional form of each LQ45 stock and measure the Mean Absolute Percentage Error (MAPE) of each stock. ARIMA (p, d, q) consists of AR(p) and MA(q) models as well as combining d differencing processes. ARIMA (p, d, q) modeling briefly contains several processing stages: parameter estimation, residual test, and prediction. This study showed that the mean forecast error using MAPE for LQ45 stock was 10.088%, with a standard deviation of 8.968%. Furthermore, BBCA stocks had the lowest forecast error of 2.1797%, and MDKA stocks had 44.49%. The highest forecast error was due to MDKA stocks having visually and exponentially increased. Therefore, it can continue increasing exponentially or even decreasing sharply in the future price period.