This project focuses on predicting the daily closing prices of Reliance Industries Ltd. using historical stock data from 2017 to 2022. It demonstrates the application of Time Series Analysis and Statistical Modeling to financial data.
- Exploratory Data Analysis (EDA): Visualized trends and volatility in stock prices.
- Stationarity Testing: Performed the Augmented Dickey-Fuller (ADF) Test.
- Transformation: Applied Log-Transformation and 1st-order Differencing to achieve stationarity.
-
Model Selection: Identified optimal parameters (
$p, d, q$ ) using ACF and PACF plots. - Final Model: Selected ARIMA(1, 1, 0) based on AIC criteria.
- Metrics: Achieved a 0.2% Mean Absolute Percentage Error (MAPE).
- Conclusion: The model provides highly accurate short-term (next-day) forecasts, making it a viable tool for baseline financial predictions.
- Pandas,
NumPy,Statsmodels,Matplotlib,Seaborn.
