Sales Forecasting Using Time Series Analysis (ARIMA / SARIMA)

Forecast monthly e-commerce sales using ARIMA and SARIMA models to support inventory management and resource allocation. Project Structure

DS3_SalesForecasting__config.py       ← All parameters
DS3_SalesForecasting__data_gen.py     ← 60-month synthetic sales generator
DS3_SalesForecasting__stationarity.py ← ADF test + ACF/PACF plots
DS3_SalesForecasting__arima_model.py  ← ARIMA fit, forecast, evaluate
DS3_SalesForecasting__sarima_model.py ← SARIMA fit, future forecast
DS3_SalesForecasting__dashboard.py    ← Raw series + forecast comparison charts
DS3_SalesForecasting__main.py         ← Entry point
DS3_SalesForecasting__requirements.txt

Run

pip install -r DS3_SalesForecasting__requirements.txt
python DS3_SalesForecasting__main.py

Results

Model MAE RMSE MAPE

ARIMA(1,1,1) ~$420 ~$530 ~5.2%

SARIMA(1,1,1)(1,1,1,12) ~$290 ~$375 ~3.6%

SARIMA captures seasonal patterns; ARIMA misses Q4 peaks

6-month forward forecast generated for inventory planning