Maximum-Likelihood parameters for Bernoulli trials with decaying success probability

140 Views Asked by At

Say I have a sequence of independent bernoulli trials $X_0, X_1, \ldots, X_n$, with exponentially decaying success probability $\mathbb{P}(X_t=1)=r^{-t}$, for some unknown parameter $r$. How can I find the maximum likelihood estimate of $r$, in terms of $x_0,x_1,...,x_n$?

What about the more general case for two unknown parameters, when $\mathbb{P}(X_t=1)=ar^{-t}, \quad0<a\le1$?

Thanks, Luke