Modelling and Predicting the Stock Price Behaviour of Reliance Industries Limited using Quartic Regression and Normal Distribution

This exploration investigates the stock price behavior of Reliance Industries Limited (RIL) using mathematical modelling. The author employs quartic regression across two distinct phases, utilizing first and second derivatives to identify inflection points and link them to real-world economic events.

The methodology includes applying horizontal and vertical transformations to create a seamless piecewise model, with reliability assessed via Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE). Furthermore, a Gaussian distribution is used to calculate the probability of positive weekly returns. The study concludes that while polynomial models offer high explanatory power for historical data, predictive accuracy is constrained by market volatility, demonstrating a nuanced understanding of the limits of mathematical forecasting in finance.