Question

ECONOMETRICS 2

a) What are the 3 conditions for a time series to be Covariance Stationary?

b) What is meant by detrending a time series variable? Assume that we detrend all variables (both the dependent and all independent variables) in a regression equation and then run an OLS estimation. What is this estimated regression equation equivalent to? How you interpret the estimated coefficient of this equation?                                                                                                                                               

Answer 1

Answer a- For a time series to be classified as stationary (covariance stationarity), it must satisfy 3 conditions:

1. Constant mean-The first property implies that the mean function must be constant.
2. Constant variance-The second property implies that the covariance function depends only on the difference between time period 1 and time period 2 and only needs to be indexed by one variable rather than two variables.
3. Constant covariance between periods of identical distance- It implies that the covariance is indepe3endent of time.