Prediksi Data Time Series Harga Penutupan Saham Menggunakan Model Box Jenkins ARIMA

The ability to predict time series data on closing market prices is critical in determining a company's stock results. The development of an efficient stock market has a positive correlation with economic growth, in a country both in the short and long term. In practice, investors tend to invest in countries that have a stable economy, low crime. The rise and fall of stock prices has made many investors develop various effective strategies in predicting stock prices in the future with the aim of making investment decisions so that investors can guarantee their profits and minimize risk.As a result, the researchers developed a model that could accurately estimate precision. Time series data models are one of the most powerful methods to render assumptions in decisions containing uncertainty. The AutoRegressive Integrated Moving Average (ARIMA) model with the Box Jenskins time series procedure is one of the most commonly used prediction models for time series results. The steps for using the Box Jenskins ARIMA model for historical details of expected stock closing prices are outlined in this paper. BBYB and YELO stock data from yahoo.finance were used as historical data. The Aikake Information Criterion (AIC), Bayesian Information Criterion (BIC) / Schawrz Bayesia Criterion (SBC), Log Probability, and Root Mean Square Error (RMSE) are used to choose an effective model, and the model chosen is ARIMA (1 , 1,2). The findings suggest that the Jenkins ARIMA box model has a lot of scope for short-term forecasting, which may help investors make better decisions. Keywords: prediction, the stock's current closing price, Box Jenskins ARIMA mode