The plots show the simulated (predicted) outputs of selected models. The models are fed with inputs from the Validation Data set, whose output is plotted in black (in white on a black blackground).
The percentage of the output variations that is reproduced by the model is displayed at the side of the plot. A higher number meas a better model.
The precise definition of the fit is:
FIT = [1 - NORM(Y - YHAT)/NORM(Y -MEAN(Y))]*100
where Y is the measured output and YHAT is the simulated/predicted model output. The time span over which the fit is measured can be changed under the Options submenu Customized time span for fit.
There are submenus under the Options menu, which allow you to choose between simulated and predicted model output. There are also options to show measured and model outputs together or to show the difference between them.
An intuitive interpretation of a K-step ahead predicted output is to see it as a result of a simulation, using the actual input, that was started at the correct (measured) output level K samples earlier. K is called the prediction horizon. It can be changed under the Options submenu Set prediction horizon.
(file compare.htm)