Which of the following methods is most suitable for forecasting future sales based on historical sales data that exhibits seasonal trends ?

Explanation: The ARIMA (AutoRegressive Integrated Moving Average) model is widely used for time series forecasting, especially when the data shows seasonal patterns or trends. ARIMA models combine three components: auto-regression (AR) , differencing (I) , and moving average (MA) . It is particularly effective for forecasting future values based on past data by capturing trends and seasonality. For datasets with clear seasonal fluctuations, a seasonal ARIMA (SARIMA) model can be used, which extends ARIMA by adding seasonal terms to capture seasonal patterns more explicitly. This makes ARIMA the most suitable method for sales forecasting when trends and seasonal variations exist. Option A: A Simple Moving Average smoothens data by averaging over a window but fails to capture seasonality and trends explicitly, making it less effective for forecasting complex seasonal data. Option B: Exponential Smoothing is useful for forecasting, especially when newer observations are more relevant, but it may not adequately model both seasonal patterns and trends as effectively as ARIMA. Option D: Linear Regression can be used for trend forecasting, but it does not handle seasonality in time series data well unless additional seasonal terms are added. Option E: Naive Bayes is a classification technique and not suitable for forecasting time series data with trends or seasonality.