Main Content

plotinerrcorr

Plot input to error time-series cross-correlation

Syntax

plotinerrcorr(x,e)
plotinerrcorr(...,'inputIndex',inputIndex)
plotinerrcorr(...,'outputIndex',outputIndex)

Description

plotinerrcorr(x,e) takes an input time series x and an error time series e, and plots the cross-correlation of inputs to errors across varying lags.

plotinerrcorr(...,'inputIndex',inputIndex) optionally defines which input element is being correlated and plotted. The default is 1.

plotinerrcorr(...,'outputIndex',outputIndex) optionally defines which error element is being correlated and plotted. The default is 1.

Examples

Plot Cross-Correlation of Inputs to Errors

Here a NARX network is used to solve a time series problem.

[X,T] = simplenarx_dataset;
net = narxnet(1:2,20);
[Xs,Xi,Ai,Ts] = preparets(net,X,{},T);
net = train(net,Xs,Ts,Xi,Ai);
Y = net(Xs,Xi,Ai);
E = gsubtract(Ts,Y);
plotinerrcorr(Xs,E)

Figure Input-Error Cross-correlation (plotinerrcorr) contains an axes. The axes with title Correlation between Input 1 and Error 1 contains 4 objects of type bar, line. These objects represent Correlations, Zero Correlation, Confidence Limit.

Introduced in R2010b