MATLAB: Time series prediction: R nearly 1 but MSE very high

Deep Learning Toolboxmean square errormserregression

Hi,

I am trying to predict future outcome of a nonlineare function using NARX.

My Regression R values are all very close to 1. 0.9998 0.9997 0.997 ….

Although my MSE values are very high, they arer about 10364

The average number of the outputs is: 5600

Thanks a lot!

Best Answer

The relationship between R and MSE is

 R = sqrt( 1 - MSE/mean(var(target',1)))

Therefore the conclusion is that the mean target variance is much greater than 10364

Hope this helps.

Thank you for formally accepting my answer

Greg

Related Solutions

MATLAB: Neural Network: what does mean when the R-value is close to 1, but MSE (mean squared error) is large

 R^2 =  1 - MSE/mean(var(target',1))

If the R value is large, then the MSE value is much smaller than the mean target variance. This indicates that the net has successfully managed to model most of the variation in the input->target transformation.

Hope this helps.

Thank you for formally accepting my answer

Greg

MATLAB: How to create a neural network with 1 layer only (no hidden layers)

 net = fitnet([]);     % For regression
 net = patternnet([]); % For classification

Hope this helps.

Thank you for formally accepting my answer

Greg

Best Answer

Related Solutions

MATLAB: Neural Network: what does mean when the R-value is close to 1, but MSE (mean squared error) is large

MATLAB: How to create a neural network with 1 layer only (no hidden layers)

Related Question