Forecasting using AR(p) model with real-time data. NEED HELP !!!!!

Hi there !

I would like to share with you my problem. I want to forecast recursively the next 12 (unknown) observations. I have attached my real-time dataset which is in a triangular format. As you can see in my excel file, I want to forecast the next 12 (unknown) observations of each column. I am using an AR(8) model. Let me also give you the code I have developed.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

[rows,columns]=size(NEW_ROUTPUTQvQ19652014FilledMissingValues);

 for column=1:columns;  
  InSampleData=NEW_ROUTPUTQvQ19652014FilledMissingValues(1:end,column);
  find(InSampleData==0);
  data=InSampleData(1:(ans-1),1);

    for periods_ahead = 1:12;
     Mdl = arima(1,0,0);
     EstMdl = estimate(Mdl,data);
     [Yhat,YMSE] = forecast(EstMdl,1);
     data(end+1,1)= Yhat;
     zeros=find(NEW_ROUTPUTQvQ19652014FilledMissingValues==0);
     pos=zeros(1,1);    
     NEW_ROUTPUTQvQ19652014FilledMissingValues(pos,column)=Yhat;
    end;
end;

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

If something is not clear or you have questions I am happy to answer you asap.

Could you please help me with it..??? Think I am desperate.!!