Consider a binomial experiment with probability of success $p$ in which $m$ trials are conducted resulting in $R$ successes. A further set of trials is then conducted until $s$ further successes have occurred. The number of trials necessary in the second set is a random variable $N$. Show that $(R,N)$ is a sufficient statistic for $p$, but not complete.
I know how to calculate a UMVUE for $p$ in a set of bernoulli trials, I reached $p^*=T/n$, where $n$ is the number of trials and $T$ is the sum of the observations. I know the definitions of sufficient, complete, UMVUE but not sure how to adapt it into something like $(R,N)$ as a statistic.