This video introduces the simple mean square error (MSE) as a criterion
which trades-off bias and variance for an estimator. The relationship
between the MSE and bias and variance is defined. The minimum MSE is
introduced as an estimator which would appear to produce an improved
design. However, through an example, it is shown that such estimators
are sometimes unrealisable if there is bias. Nevertheless, there are
applications where the MSE can produce results, or indeed inspire other
estimators, such as estimators for variance (examples will be given
later in the course).
PGEE11164 Probability, Estimation Theory, and Random Signals Lectures -- School of Engineering, University of Edinburgh. Copyright James R. Hopgood and University of Edinburgh, Scotland, United Kingdom (UK). 2020.
Institute for Digital Communications, Alexander Graham Bell Building, The King's Buildings, Thomas Bayes Road, Edinburgh, EH9 3JL. UK.