Ameya Deoras

Izi, that error seems to suggest that you are calling the genPredictors function without any inputs. “data” is a required input for the function and needs to be specified. See loadScriptNN for an example of how to use this function.

Jose, this function won't eliminate the dead space you see because it only modifies the ticks that would be drawn by datetick. What you need is the ability to modify the x-values in your dataset so that when they are plotted the data are no longer evenly spaced in the x-dimension. This is a much more challenging problem that this function is not set up to address.

Here is a function that can be used to run multi-step prediction from a 1-step feed-forward network:

function pred = multiPredict(net, X, numSteps)
% multiPredict performs iterated prediction with a feed-forward neural
% network.
%
% SYNTAX: pred = multiPredict(net, X, numSteps)
%
% INPUTS:
% net is a feed-forward neural net
% X is a matrix of size numObservations * numPredictors
% numSteps is the number of steps to predict (eg. 10)
%
% OUTPUTS:
% pred is a matrix of size numObservations * numSteps where every column is
% the ith step prediction for that sample observations. The first column is
% the 1-step prediction, the second columns is the 2-step prediction and so
% on.

numObs = size(X,1);
pred = zeros(numObs, numSteps);
pred(:, 1) = net(X')';
for i = 2:numSteps
X = [pred(:,i-1) X(:,1:end-1)];
pred(:,i) = net(X')';
end

Excellent script. I have a question. Is there a way to color only certain values? Say, between 0 and 0.001 have one color, between 0.001 and 0.01 a different color, between 0.01 and 0.05 a different color, and everything else appear as white/blank? Thanks for your help.

Dear Ameya,

In order to forecast the electricity price of the day ahead Dutch market I am using neural networks and real data from 2007-2011.
I constructed the X matrix as you said in the webinar for load forecasting and proved with different combinations. The one with gives me the lower MAPE value consists of 8 inputs (Hour dayOfWeek isWorkingDay prevWeekSameHourPrice prevDaySameHourPrice prev24HrAvePrice prevDayNGPrice prevWeekAveNGPrice), but I can't reduce this value more than 11%.

I need to achieve better accuracy but I don't know whar should I change. Could you send me an example of who to forecast the day ahead electricity price? My email is fran90_7@hotmail.com

Thank you in advance.

Kind regards,

Jose Francisco Bolado