Plot confidence interval for the slope

Hi all,

I've been looking for an answer for a while, but it seems nobody asked it before.

So : I have a linear regression. I can find the confidence interval of the slope with the regress function. It gives me 2 numbers (for the slope). Of course. But how to plot it ? I mean, plots of the "confidence interval of the slope" look like curves when I google it. How do I obtain those curves ?

Thank you very much for your help !

Lucile

Answer 1

[Wayne King]

You will get more than the confidence interval of the slope. In a simple linear (1st order) regression model, keep in mind that you will have a confidence interval for the intercept and the slope terms. You can plot those lines as confidence bounds, see the following example

 load carsmall
 x1 = Weight;
 y = MPG; 
 X = [ones(size(x1)) x1];
 [b,bint] = regress(y,X) 
 xval = min(x1):0.01:max(x1);
 yhat = b(1)+b(2)*xval;
 ylow = bint(1,1)+bint(2,1)*xval;
 yupp = bint(1,2)+bint(2,2)*xval;
 plot(x1,y,'k*');
 hold on;
 plot(xval,ylow,'r-.');
 plot(xval,yupp,'r-.');
 plot(xval,yhat,'b','linewidth',2);

[Lucile]

Thank you very much for your answer ! Problem solved ! (and I feel stupid)