1
$\begingroup$

consider this equation 1: \begin{equation} \label{eq:1} C_{t} = \beta_{1} + \lambda C_{t-1} + \epsilon_{t} \end{equation} If the error term is independently and identically distributed (iid) with mean 0 and constant variance and \begin{equation} \label{eq:2} E(\epsilon_{t}\mid C_{t-1}) = 0 \end{equation} i.e., the error is not correlated with \begin{equation}C_{t-1})\end{equation} Is the OLS estimator of the coefficients in equation 1 unbiased and consistent under this new error specification? Why?

My answer: I would assume unbiased and consistent because the expectation of the errors and the lagged variable = 0 ? Thoughts?

$\endgroup$

1 Answer 1

2
$\begingroup$

OLS is consistent but biased. For consistency, we need $Cov(\epsilon_t, C_{t-1})=0$. The condition, $E[\epsilon_t|C_{t-1}]=0$ is sufficient for the covariance to be 0.

For unbiasedness, we need $E[\epsilon_t|C]=0$ where $C$ is the vector of $C_t$ at all time periods.

$\endgroup$
2
  • $\begingroup$ Just a follow up: Consider the following regression equation (* ) \begin{equation} \label{eq:1} C_{t} = \beta_{1} + \lambda C_{t-1} + \epsilon_{t} \end{equation} and let \begin{equation} \label{eq:3} \epsilon_{t} = \rho\epsilon_{t-1} + u_{t} \end{equation} where the error component \begin{equation} u_{t}\end{equation} is iid with mean 0 and constant variance, and \begin{equation} \label{eq:4} E(u_{t}\mid C_{t-1},\epsilon_{t-1}) = 0 \end{equation} Is the OLS estimator of the coefficients in (*) unbiased and consistent under this error specification? Why? $\endgroup$ Apr 23, 2022 at 14:24
  • $\begingroup$ OLS will be biased for the same reason, although it may be consistent. I responded on your other question. $\endgroup$ Apr 25, 2022 at 8:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.