In this example, we see how it's possible to apply the EM algorithm for the estimation of unknown parameters (inspired by an example discussed in the original paper Maximum likelihood from incomplete data via the em algorithm, Dempster A. P., Laird N. M., Rubin D. B., Journal of the Royal Statistical Society, B, 39(1):1–38, 11/1977).
Let's consider a sequence of n independent experiments modeled with a multinomial distribution with three possible outcomes x1, x2, x3 and corresponding probabilities p1, p2 and p3. The probability mass function is as follows:
Let's suppose that we can observe z1 = x1 + x2 and ...