The EM algorithm is a method for producing a sequence of parameter est
imates that, under mild regularity conditions, converges to the MLE. T
he EM algorithm is well regarded, in part because of two monotonicity
properties: convergence to the MLE is monotone, and the value of the L
ikelihood function increases with each iteration. A graphical illustra
tion of the EM algorithm makes these properties intuitively apparent i
n the one-parameter case. In addition, a well-known result regarding t
he rate of convergence of the algorithm can be inferred.