2
$\begingroup$

I have the pdf

$$f(y ; \theta) = \frac{1}{\theta} \exp( \frac{-y}{\theta}), \ y > 0$$

and I'm supposed to determine if the following two estimators are unbiased or not: $ \hat \theta = nY_{min} $ and $ \hat \theta = \frac{1}{n}\sum_{i=1}^n Y_i $. I'm running into some problems because when I try to find the expected value of $ Y_{min} $ and $ Y_i $, the integral is undefined.

$\endgroup$

2 Answers 2

2
$\begingroup$

Please see the following Wikipedia article and it shows that both the integral exist. Also, if you look lower in the article, you will also see that both the estimators are unbiased.

$Y_{min}$ is also exponentially distributed with mean parameter $\frac {\theta} {n}$

Therefore, $E[nY_{min}]=n\times\frac {\theta} {n} = \theta$. Similarly, $E[\frac{1}{n}\sum_{i=1}^n Y_i] = \frac{n \theta}{n}=\theta$

I am sure both the integrals exists

$\endgroup$
1
  • 1
    $\begingroup$ Ya, realized what I was doing. Missed that whole exponential thing. Thanks. $\endgroup$
    – tshauck
    Commented Feb 9, 2011 at 4:28
0
$\begingroup$

The fact that the sample mean is an unbiased estimator is obtained combining these two facts: 1. The sample mean is an unbiased estimator of the population mean 2. The population mean is equal to theta

$\endgroup$

Your Answer

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

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