I'm trying to teach myself Bayesian statistics and am currently trying find the posterior distribution on the following problem:

Suppose $X_1,...,X_n$ are iid exponential random variables, with density $f(x;\theta) =\theta e^{-\theta x}$ ,and let us suppose that we have a prior on $\theta$ with density $\theta^{\alpha -1}e^{-\beta \theta}$

Now, to find the posterior, we do Likelihood times Prior. The likelihood is given by

$$\theta^n e^{-n^2\theta \bar{x}}$$

So that we have

$$\theta^{\alpha -1}e^{-\beta \theta}\theta^n e^{-n^2\theta \bar{x}}=\theta^{n+\alpha-1}e^{-n^2 \theta \bar{x} - \beta \theta}$$

And in the textbook I am reading, from here you are meant to deduce the posterior distribution, but I cannot see an obvious distribution. Have I made an error? Can the posterior density function contain the random variable (x in this case)?

  • 1
    $\begingroup$ The posterior on $\theta$ is the conditional on $x$, so it is natural to find $x$ in the density. $\endgroup$ – Xi'an Oct 18 at 5:57
  • $\begingroup$ @Xi'an do we still treat $x$ as a random variable in this case? Has it been observed as the posterior is conditional on $x$? $\endgroup$ – Bill Oct 18 at 9:06

Your likelihood should have $n$ instead of $n^2$. Anyway, the function you found is proportional to the pdf of a Gamma Distribution.

| cite | improve this answer | |

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.