Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Let $X1, \dots, Xn$ be a random sample of size n from the continuous distribution with pdf $f_X(x|\theta) = \frac{e^{-x}}{1-e^{-\theta}} I(x)_{[0,\theta]} I(\theta)_{(0, \infty)}$.

(1) Find the maximum likelihood estimator for $\theta$.

(2) Find the maximum likelihood estimator for the median,$\lambda$, of this distribution.

For (1), I got $X_{(n)}$ because I thought if $\theta$ was minimized, it would maximize the function.

For (2), I know I set up the distribution to solve for the value of the median. I think that once I find this value I will use it in relation to the MLE from (1) to find the MLE for(2). I got $m = ln(2) - ln(e^{-\theta} + 1)$ but this doesn't seem right and I amd not sure where I am messing up. Any assistance is greatly appreciated.

share|cite|improve this question
up vote 2 down vote accepted

Consider $s_n=x_1+\cdots+x_n$ and $t_n=\max\{x_k\mid1\leqslant k\leqslant n\}$, then $$ f(x_1,\ldots,x_n\mid\theta)=\mathrm e^{-s_n}(1-\mathrm e^{-\theta})^{-n}\mathbf 1(\theta\geqslant t_n). $$ Since $2\mathrm e^{-\lambda}=1+\mathrm e^{-\theta}$, this is also $$ f(x_1,\ldots,x_n\mid\theta)=\mathrm e^{-s_n}2^{-n}(1-\mathrm e^{-\lambda})^{-n}\mathbf 1(\lambda\geqslant\log2-\log(1+\mathrm e^{-t_n})). $$ Thus, $f(x_1,\ldots,x_n\mid\theta)$ is indeed maximum when $\lambda=\hat\lambda$, with $$ \hat\lambda=\log2-\log(1+\mathrm e^{-t_n}). $$ Note that $\hat\lambda=g(\hat\theta)$ where the function $g$ is defined by $\lambda=g(\theta)$.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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