How can you capture seasonal fluctuations in time series analysis?

• A. Use a yearly moving average.
• B. Apply a log transformation.
• C. Ignore seasonality.
• D. Use a 3-month moving average.

Answer: (D)

Seasonality shows up as repeating patterns at fixed intervals, so to reveal that pattern you want smoothing that matches the length of the seasonal cycle. If the data have a three-month (quarterly) cycle, applying a three-month moving average smooths over adjacent months and lets the recurring quarterly highs and lows stand out. This makes it easier to estimate the seasonal component or seasonal indices by comparing the original values to the smoothed baseline. A yearly moving average would blur those within-year fluctuations, a log transform changes variance but not the seasonal pattern, and ignoring seasonality won’t capture the repeating effects. So, using a three-month moving average aligns with the season length and best captures seasonal fluctuations.