Predicting sales during COVID using Machine Learning Techniques

The purpose of this study is to compare VAR, ARIMA and SARIMA methods in an attempt to generate sales forecasting in Store xyz with high accuracy. This study will compare the results of sales forecasting with time series forecasting model of Vector Auto Regression (VAR), Autoregressive Integrated Moving Average (ARIMA) and Seasonal Autoregressive Integrated Moving Average (SARIMA). VAR or ARIMA model still accurate when the time series data is only in a short period, these models is accurate on short period forecasting but less accurate on long period forecasting. Meanwhile Seasonal Autoregressive Integrate Moving Average is more accurate on forecasting seasonal time series data, either it’s pattern shows trend or not all three models are compared with forecasting data showing seasonal patterns. The data used is the data of super mart retail store, sales from 2017 to 2022. Accuracy level of each model is measured by comparing the percentage of forecasting value with the actual value. This value is called Mean Absolute Deviation (MAD). Based on the comparison result, the best model with the smallest MAD value is SARIMA model (0,1,0) (0,1,0)12 with MAD value 0.122. From the comparison results can be concluded that the SARIMA model is optimal to be used as a model for further forecasting Keywords: Machine Learning, sales prediction, ARIMA, SARIMA, VAR, PYTHON, Anaconda navigator, Jupiter notebook.