Skip to main content

Parameter Mean Squared Error

Edited by: Published: 2018
+- LessMore information
Download PDF

The parameter mean squared error (MSE), also known as empirical mean squared error, indicates the deviation of an estimated value from the expected value of a given parameter. The lower the MSE is, the better accuracy an estimated value or an estimation method presents. Mathematically, it is formulated as the average of the squared deviations across a certain number of estimations; thus, the MSE is always a positive value. To calculate the MSE, one needs to know the expected values of the parameters, which normally are unknown in statistical analysis. For this reason, the MSE is commonly used as an evaluation criterion in conjunction with the Markov chain Monte Carlo (MCMC) method, in which data are randomly sampled from probability distributions rather than collected from ...

Looks like you do not have access to this content.

Reader's Guide