Skip to content

This repository provides a comprehensive solution for tractor sales forecasting, implementing and evaluating SARIMAX and Exponential Smoothing models to identify the most accurate predictive approach.

Notifications You must be signed in to change notification settings

Kishan-Sinha/Demand-Forecasting-Using-Time-Series-Analysis

Repository files navigation

Demand-Forecasting-Using-Time-Series-Analysis

Time Series Forecasting of Tractor Sales: A comparative analysis and model recommendation using SARIMAX and Exponential Smoothing. It's main components are:

1. Data preparation, modeling and evaluation: This Jupyter Notebook details the complete time series forecasting workflow for tractor sales. It covers data loading, preprocessing, exploratory data analysis (EDA) including visualization of sales trends, stationarity checks (rolling statistics, seasonal decomposition, ACF/PACF plots), and the implementation of two key forecasting models: SARIMAX and Exponential Smoothing. The notebook concludes with the in-sample prediction and evaluation of these models using metrics such as MAE, MSE, RMSE, and MAPE

2. Evaluation Summary: It includes a table summarizing key performance metrics for both models. Based on this quantitative analysis, a clear recommendation for the optimal forecasting model is provided, along with a justification of the choice

3. Tractor Sales.csv: The original Raw file on which the Whole analysis was carried on. It contains a structured sales record of tractors from an US based manufacturer

About

This repository provides a comprehensive solution for tractor sales forecasting, implementing and evaluating SARIMAX and Exponential Smoothing models to identify the most accurate predictive approach.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published