Time Series Analysis DATA-SCIENCE-FOUNDATIONS Practice Questions (Page 2)

Questions 6–10

1. 6application · medium

   A data analyst is fitting an ARIMA model to a time series that required first differencing to become stationary. The analyst wants to forecast future values. Which component of the ARIMA model accounts for the differencing applied to the original series?

   • AThe 'd' parameter in ARIMA(p,d,q)
   • BThe seasonal component in a SARIMA model
   • CThe 'q' parameter in ARIMA(p,d,q)
   • DThe 'p' parameter in ARIMA(p,d,q)
   Reveal Solution Discussion Select an answer first
2. 7foundation · easy

   In time series decomposition, which component represents the long-term direction of the data, typically over many years, and is not necessarily a fixed pattern?

   • AIrregular
   • BSeasonality
   • CCyclical
   • DTrend
   Reveal Solution Discussion Select an answer first
4. 8foundation · easy

   What is the main purpose of seasonal decomposition of a time series?

   • ATo identify outliers in the data.
   • BTo make the series stationary by removing the mean.
   • CTo convert the series into a normal distribution.
   • DTo separate the series into trend, seasonal, and residual components for analysis.
   Reveal Solution Discussion Select an answer first
5. 9foundation · easy

   Which statement correctly defines a stationary time series?

   • AIts mean changes over time, but the variance remains constant.
   • BIt has no trend or seasonality, but the variance can change over time.
   • CIts values are independent and identically distributed with no autocorrelation.
   • DIts mean and variance are constant over time, and the covariance between two time points depends only on the lag between them.
   Reveal Solution Discussion Select an answer first
6. 10application · medium

   A data scientist is preparing to fit an ARIMA model to a company's weekly sales data. The time series plot shows a clear upward trend and the variance increases with the level of the series. The data scientist wants to apply ARIMA, which assumes stationarity. What should the data scientist do first?

   • AFit the ARIMA model directly to the raw data because ARIMA can handle trends
   • BApply a log transformation and then first-difference the series
   • CRemove the seasonal component using seasonal decomposition and then fit the model
   • DUse exponential smoothing instead because it does not require stationarity
   Reveal Solution Discussion Select an answer first