Upgrade subject to access all content
What is a consequence of serial autocorrelation?
The expected value of the disturbance term is $E[\varepsilon_t] \neq 0$.
The OLS parameter values are biased.
Typical t-statistic and F-statistic are valid for making inferences.
The Gauss-Markov assumptions are not violated.
The covariance between an error term in time $t$ and an error term in time $s$ is $Cov(\varepsilon_t,\varepsilon_s)\neq 0$ for $t\neq s$.