How do I find a least squares regression line of 4 points in a plane?

How would I find the least squares regression line for the 4 given (x,y) points {{-3, 0}, {-1, 1}, {0, 2}, {2, 3}}? I am trying to use the LeastSquares command, but not having much success understanding the documentation and thus getting the command to work properly.

Thanks,

Mitch Sandlin

You should use LinearModelFit rather than LeastSquares given that you have the data points rather than the structure m.x == b as described in the documentation.

Or this:

data = {{-3, 0}, {-1, 1}, {0, 2}, {2, 3}};
line = Fit[data, {1, x}, x];
Show[Plot[line, {x, -3.5, 2.5}], ListPlot[data, PlotStyle -> Red], PlotLabel -> ToString[line]]

Cheers,

M.