Pseudocode for first-order least-squares calculation.

Assumptions: the data are in the arrays x and y;
length(x) returns the nunber of points; 
x(i) is the ith element of array x, where i is integer;
x^2 is the square of x, etc.
The for loops (for i=1:n....end) increment i in unit steps.

n=length(x)
sumx=0
sumy=0
sumxy=0
sumx2=0
for i=1:n
   sumx=sumx+x(i)
   sumy=sumy+y(i)
   sumxy=sumxy+x(i)*y(i)
   sumx2=sumx2+x(i)*x(i)
end
meanx=sumx/n
meany=sumy/n
slope=(n*sumxy-sumx*sumy)/(n*sumx2-sumx*sumx)	
intercept=meany-(slope*meanx)
ssy=0
ssr=0;	
for i=1:n
   ssy=ssy+(y(i)-meany)^2
   ssr=ssr+(y(i)-intercept-slope*x(i))^2
end
R2 = 1-(ssr/ssy)