Mean quadratic error

**user** · August 15th 2010, 09:24 AM

To prove that mean quadratic error of a parameter is the sum of varianca of parameter and quadratic B of parameter. Thanks

**SpringFan25** · August 15th 2010, 02:34 PM

Did you mean estimator instead of parameter? Outside of bayesian statistics, parameters do not normally have a variance.

Estimate: X
True Value: Y

$E(X-Y)^2 = E(X^2) - 2E(XY) + E(Y)^2$

$E(X-Y)^2 = E(X^2) - 2YE(X) + Y^2$

$E(X-Y)^2 = E(X^2) - 2YE(X) + Y^2 + E(X)^2 - E(X)^2$

$E(X-Y)^2 = E(X^2) - E(X)^2 - 2YE(X) + Y^2 + E(X)^2$

$E(X-Y)^2 = Var(X) + Y^2 - 2YE(X) + E(X)^2$

You may find it interesting to note that the quatratic ("B") factorises to give

$E(X-Y)^2 = Var(X) + E(Y - E(X))^2$

So that for an unbiased estimator, the MSE is equal to the variance of the estimate (as you'd expect from the definition of a variance).

Thread: Mean quadratic error

LinkBack

Thread Tools

Display

Mean quadratic error

Similar Math Help Forum Discussions

best fit, minimizing point-to-line error, not y error

Relative Error & Maximum Error

Can someone please tell me my code wont work, error after error

Linear least square approximation error estimation with error in both coordinates

Integration of quadratic denominaters error

Search Tags