Explanation

Correct Answer: D

A moving average representation shows evidence of autocorrelation cutoff. This means that the correlation between a variable and its lagged values abruptly drops to zero after a certain point. This is a key characteristic of moving average models.

Why the other options are incorrect:

• A: Autoregressive processes can be covariance stationary under certain conditions (when the roots of the characteristic equation lie outside the unit circle).
• B: This is actually the opposite - autoregressive processes show gradual decay in autocorrelations, not cutoff.
• C: This statement is reversed - moving average processes show autocorrelation cutoff, while autoregressive processes show gradual decay.

Key Distinction:

• Moving Average (MA) Process: Autocorrelation function cuts off after q lags (for MA(q) process)
• Autoregressive (AR) Process: Autocorrelation function decays gradually (exponentially or sinusoidally)

This difference is crucial in time series analysis for identifying the appropriate model type based on the observed autocorrelation pattern in the data.