getTrend,forcast, and Simulate timeseries data in matlab?

I have climate data of four different stations of past 30 years i want to predict/forecast this data for next 5 years. This climate data is in data matrix of size 31*4 (30 years and four stations).

I am trying to forecast, getting trend and simulate this model based on this 30-year time series data to next five-year using ar model.

Here is my just effort(not test yet)

data=fineM; %is a matrix of size 31 by 4
for i=1:length(data,2)
    mysys(:,jj)=ar(fineM(:,jj),2); %creating a matrix of ar model
end
for j=1:size(data,2)
T(:,j) = getTrend(data(:,i),1); %preparing matrix of gettrend of size 31*4
F(:,j) = forecast(mysys(j),data(j),5); % trying to forcast for next five years
mysimu=sim(mysys(j),data(j)); % trying to simulate 
end

I am totally new to matlab, and even confuse with these terms forcast, getTrend,retrend,predict,sim. Any expert please guide me please how can i forcast my values for next five years also make it a simulated model so that i can check my accuracy of model.

[Muhammad Usman Saleem]

i am not sure about looping

[G Newton]

dear Muhammad ;

I am totally new to these functions not know what i am doing?