Question 6
A data scientist is building a time series model to forecast monthly sales for a retail store. The dataset consists of 5 years of historical sales data. Before fitting the model, they observe the following:
1. There is a clear upward trend in sales over the years. 2. Sales exhibit seasonality, with peaks during the holiday season each year. 3. The variance in sales increases as the trend rises.
Based on these observations, which of the following preprocessing or modeling steps should be considered?
Apply differencing to remove trend.
Use a Seasonal ARIMA (SARIMA) model to capture trend and seasonality.
Apply log transformation to stabilize variance.
Check the stationarity of the series before modeling.
All of these.