0
$\begingroup$

$ E(\hat \theta_1 ) = E(\hat \theta_2) = \theta_1 $ , $ Var (\hat \theta_1) = \sigma_1 , Var(\hat \theta_2) = \sigma_2, Cov(\hat \theta_1, \hat \theta_2) = \sigma_{12}$

$\hat \theta_3 = a \hat \theta_1 + (1-a) \hat \theta_2 $

The question asks me to find the value of a that minimizes the variance. I have the solutions to the answer but I don't understand how the covariance is used in calculating $V(\hat \theta_3) $

$V(\hat \theta_3) $ = $ V(a \hat \theta_1 + (1-a) \hat \theta_2 ) = $ $a^2 V( \hat \theta_1) + (1-a)^2 V(\hat \theta_2) + 2(1-a)Cov(\hat \theta_1, \hat \theta_2)$

My issue isn't with determining the minimum value of alpha, that's just found by setting the derivative of $V(\hat \theta_3)$ to $0$ and checking 2nd deriviative for if it is a minimum, my issue is with the $(1-a)$

I do know that since both $\hat \theta_1, \hat \theta_2$ are not independent, there exists a covariance $> 0$, hence obviously in the calculations of adding 2 variances, you must take into account covariance and hence the theorem;

$Var( X + Y ) = $Var(X) + Var (Y) + 2 Cov(X,Y)

but where is the (1-a) coming from in $2(1-a)Cov(\hat \theta_1, \hat \theta_2)$ ?

$\endgroup$
0
$\begingroup$

\begin{align} V(a\hat{\theta}_1+(1-\theta_1)\hat{\theta_2})&=a^2V(\hat{\theta_1})+(1-a)^2V(\hat{\theta}_2)+2Cov(a\hat{\theta}_1, (1-a)\hat{\theta}_2)\\ &=a^2V(\hat{\theta_1})+(1-a)^2V(\hat{\theta}_2)+2a(1-a)Cov(\hat{\theta}_1, \hat{\theta}_2)\\ \end{align}

I believe it's a mistake, it should be $a(1-a)$.

$\endgroup$
  • $\begingroup$ yea that would make more sense. I think it is a mistake because I noticed some other errors in these solutions I was given. Thank you! $\endgroup$ – ajdawg Feb 8 at 2:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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