Minimum Mean Squared Error Model Averaging in Likelihood Models

Ensemble averaging and mean squared error

In fields such as climate science, it is common to compile an ensemble of different simulators for the same underlying process. It is a striking observation that the ensemble mean often out-performs at least half of the ensemble members in mean squared error (measured with respect to observations). In fact, as demonstrated in the most recent IPCC report, the ensemble mean often out-performs all...

Minimum mean-squared error covariance shaping

This paper develops and explores applications of a linear shaping transformation that minimizes the mean squared error (MSE) between the original and shaped data, i.e., that results in an output vector with the desired covariance that is as close as possible to the input, in an MSE sense. Three applications of minimum MSE shaping are considered, specifically matched filter detection, multiuser ...

Mean Squared Error and Comparing Models

This criterion should be contrasted with the RSS encountered earlier in the course. The RSS pertains to model estimation, since we are already assuming a given model for some particular data set; and it suffices to estimate the specific values of our estimators for the unknown parameters. The MSE combines the previous two criteria, on the unbiasedness and the variance of β̂, through the followin...

Minimum mean squared error equalization using a priori information

A number of important advances have been made in the area of joint equalization and decoding of data transmitted over intersymbol interference (ISI) channels. Turbo equalization is an iterative approach to this problem, in which a maximum a posteriori probability (MAP) equalizer and a MAP decoder exchange soft information in the form of prior probabilities over the transmitted symbols. A number...

The Estimation Problem of Minimum Mean Squared Error