Forecasting New York Stock Exchange Trends: ARIMA in Action

A thorough time series analysis and stock price forecasting model utilizing data from the New York Stock Exchange (NYSE) is presented in this study. We make use of a sizable dataset that comprises daily stock prices in addition to important financial indicators including assets, liabilities, revenue, and income. To deal with missing values, eliminate duplicates, and guarantee date format uniformity, our method entails preparing the data. While seasonal decomposition finds long-term patterns and seasonality in the data, exploratory data analysis (EDA) exposes important stock price movements. For stock price we employ the ARIMA (Auto-Regressive Integrated Moving Average) for forecasting model. The auto-arima function was used to generate the ideal model parameters, and the Augmented Dickey-Fuller (ADF) and KPSS tests were used to assess the data's stationarity. The prediction ability of the model was evaluated using MAE and RMSE, and the ARIMA model produced low error rates (MAE: 0.022, RMSE: 0.029). Our methodology demonstrated usefulness for traders and financial experts by accurately predicting future stock values. By providing a data-driven method for anticipating stock price changes, this study advances the expanding area of financial forecasting and eventually helps investors make more informed decisions.