- CFA Exams
- CFA Level I Exam
- Topic 1. Quantitative Methods
- Learning Module 5. Time-Series Analysis
- Subject 5. Seasonality in Time-Series Models
CFA Practice Question
What is the approach to check for seasonality of a time series?
II. Test whether the residuals have significant serial correlation by using the Durbin-Watson statistic.
III. Examine the data to see whether the seasonal autocorrelations of the residuals from an AR model are significant and whether the autocorrelations before and after the seasonal autocorrelations are significant.
I. Graph the data and check for regular seasonal patterns.
II. Test whether the residuals have significant serial correlation by using the Durbin-Watson statistic.
III. Examine the data to see whether the seasonal autocorrelations of the residuals from an AR model are significant and whether the autocorrelations before and after the seasonal autocorrelations are significant.
Correct Answer: I and III
To correct for seasonality, add seasonal lags to your AR model. II is used to decide which AR model to use.
User Contributed Comments 2
| User | Comment |
|---|---|
| MasterD | The Durbin-Watson statistic is used to test Serial Correlation (CFA L2, 2008 Vol 1 Pg 301) |
| akirchner1 | Durbin -Watson can't be used for time series when the independent variable includes past values of the dependent variable. |