Time Series Analysis DATA-SCIENCE-FOUNDATIONS Practice Questions (Page 1)

Questions 1–5

1. 1application · medium

   An economist is analyzing quarterly GDP data over 20 years. The series shows a long-term upward movement, a regular pattern that repeats every four quarters, and occasional shocks due to economic crises. The economist wants to identify the long-term growth rate while ignoring the short-term fluctuations. Which component of the time series should the economist focus on?

   • ACyclical component
   • BTrend component
   • CSeasonal component
   • DIrregular component
   Reveal Solution Discussion Select an answer first
2. 2application · medium

   A hotel chain records daily occupancy rates and notices a strong seasonal pattern that repeats every year, with peaks during summer and winter holidays. The chain also observes a gradual increase in occupancy over the years. The chain wants to remove the seasonal effect to analyze the underlying trend. Which method should they use?

   • ASeasonal decomposition and then subtract the seasonal component from the original series
   • BFit an ARIMA model and use the residuals as the trend
   • CApply a 12-month moving average to the data
   • DUse exponential smoothing with a high alpha
   Reveal Solution Discussion Select an answer first
4. 3foundation · easy

   After fitting a time series model, what does it mean if the residuals resemble white noise?

   • AThe model is underfitted and needs more terms.
   • BThe model has captured the underlying patterns, and the residuals contain no remaining autocorrelation.
   • CThe data is non-stationary.
   • DThe model is overfitted and should be simplified.
   Reveal Solution Discussion Select an answer first
5. 4application · medium

   A logistics manager needs to forecast daily shipment volumes for the next week. The data shows no clear trend or seasonality but has random fluctuations. The manager wants a simple method that gives more weight to recent observations. Which forecasting method should the manager use?

   • AHolt's linear trend method
   • BSimple moving average over the last 30 days
   • CSimple exponential smoothing with a high smoothing constant
   • DSeasonal naive forecast using last year's same week
   Reveal Solution Discussion Select an answer first
6. 5application · medium

   A data analyst is checking whether a time series is stationary before modeling. The analyst computes the autocorrelation function and finds that the autocorrelations decay slowly and remain significant for many lags. What does this pattern suggest about the series?

   • AThe series is likely non-stationary and may need differencing
   • BThe series is white noise
   • CThe series is stationary and suitable for ARIMA modeling
   • DThe series has a strong seasonal component
   Reveal Solution Discussion Select an answer first