Likelihood Ratio Testing under Measurement Errors

We consider the likelihood ratio test of a simple null hypothesis (with density f 0 ) against a simple alternative hypothesis (with density g 0 ) in the situation that observations X i are mismeasured due to the presence of measurement errors. Thus instead of X i for i = 1 , … , n , we observe Z i = X i + δ V i with unobservable parameter δ and unobservable random variable V i . When we ignore the presence of measurement errors and perform the original test, the probability of type I error becomes different from the nominal value, but the test is still the most powerful among all tests on the modified level. Further, we derive the minimax test of some families of misspecified hypotheses and alternatives. The test exploits the concept of pseudo-capacities elaborated by Huber and Strassen (1973) and Buja (1986). A numerical experiment illustrates the principles and performance of the novel test.

Unscented Kalman Filter for Unobservable Parameter Estimation in Heart Cell Signals

One interesting feature of biological systems is that minor subcellular changes can cause alterations at the whole organ level. In the heart, the random dynamics of cell membrane ion channels contributes to beat-to-beat repolarization variability, which has been related to proarrhythmic risk. Inference of unobservable cellular parameters, such as the number of channels, is key to characterize such random ion channel dynamics. In this work, a methodology based on the use of Unscented Kalman Filters is proposed to infer the number of channel from action potential signals, like those commonly recorded experimentally.