Explanation

For a time series to be covariance stationary, three conditions must be satisfied:

1. Constant mean: The expected value of the time series must be constant and finite in all periods
2. Constant variance: The variance of the time series must be constant and finite in all periods
3. Constant covariance: The covariance between values at different time periods depends only on the lag between them, not on the actual time period

Therefore, both the variance and expected value must be constant and finite, making option C the correct answer.