womzhao / product-sales-forecasting

Forecasted product sales using time series models such as Holt-Winters, SARIMA and causal methods, e.g. Regression. Evaluated performance of models using forecasting metrics such as, MAE, RMSE, MAPE and concluded that Linear Regression model produced the best MAPE in comparison to other models

Product Sales Forecasting using Quantitative Methods

In a personal endevaour to learn about time series analysis and forecasting, I decided to reserach and explore various quantitative forecasting methods. The notebook in this repository documents the methods that can be applied to forecast product sales, along with a detailed explaination of the different metrics used to evaluate the forecasts.

Goal: The goal of this project was to apply various quantitative methods, (i.e. Times Series Models and Causal Models) to forecast the sales of the products available in the dataset obtained from Kaggle.

Models covered in the notebook include:

1. Seasonal Naive Model
2. Holt-Winters Model (Triple Exponential Smoothing)
3. ARIMA and Seasonal ARIMA Models
4. Linear Regression Model