Reliable parameter estimation for state space time model with repeated experiments under outlier contamination

The main purpose of this note is to use the expectation maximization algorithm  to construct a non-sensitive to outliers maximum likelihood estimator
for the linear state space invariant model with repeated measurements.
We use the Box-and-Whiskers Plot, a high breakdown but low efficiency estimator, to detect outlier observations.
To improve the efficiency of the proposed algorithm, we apply the Ranked Set Sampling method when the experiment can be replicated at a low cost. We illustrate the performance of the proposed techniques using Monte Carlo simulation.