Time Series Analysis DATA-SCIENCE-FOUNDATIONS Practice Questions (Page 3)

Questions 11–15

1. 11foundation · easy

   Why is stationarity important in time series analysis?

   • AIt eliminates the need for data cleaning.
   • BIt guarantees that the forecast will be perfectly accurate.
   • CIt ensures that the data is normally distributed.
   • DIt allows the use of statistical models that assume stable properties over time, making forecasts more reliable.
   Reveal Solution Discussion Select an answer first
2. 12foundation · easy

   Which information criterion is commonly used to compare the fit of different time series models, balancing goodness-of-fit with model complexity?

   • AAkaike Information Criterion (AIC)
   • BR-squared
   • CMean Absolute Error (MAE)
   • DP-value of the trend coefficient
   Reveal Solution Discussion Select an answer first
4. 13application · medium

   A retail analyst is examining monthly sales data for a chain of stores over the past five years. The data shows a clear upward trend, a repeating pattern that peaks every December, and a few unusual spikes during promotional events. The analyst wants to separate the underlying growth from the repeating pattern to better understand the store's baseline performance. Which approach should the analyst use?

   • AApply seasonal decomposition to break the series into trend, seasonal, and residual components
   • BCompute the moving average over a 12-month window and compare it to the raw data
   • CCalculate the autocorrelation function and inspect the lags at 12-month intervals
   • DFit an ARIMA model to the raw data and examine the residuals
   Reveal Solution Discussion Select an answer first
5. 14expert · hard

   An analyst is examining a time series that shows a clear upward trend and a regular pattern that repeats every 12 months. The analyst wants to test whether the series is stationary. Which approach is most appropriate?

   • ARemove the trend and seasonal components, then test the residuals for stationarity
   • BPerform an Augmented Dickey-Fuller test on the raw data
   • CFit an ARIMA model and check if the residuals are white noise
   • DCompute the autocorrelation function and look for a slow decay
   Reveal Solution Discussion Select an answer first
6. 15foundation · easy

   In an ARIMA(p,d,q) model, what does the 'd' parameter represent?

   • AThe degree of differencing applied to make the series stationary.
   • BThe number of autoregressive terms.
   • CThe number of moving average terms.
   • DThe seasonal period of the data.
   Reveal Solution Discussion Select an answer first