Use the return value of of the function lscov for weighted least squares?

After looking at the matlab example using the function lscov for the weighted least squares type, I was trying to use the return values b,se_b and mse? Is it possible to plot this on current figure? My assumption is that the weighted least square should be a best fit line through the data x1,y and x2,y. My ultimate goal is to find the error terms when modeling heteroscedastic data.

    x1 = [.2 .5 .6 .8 1.0 1.1]';
    x2 = [.1 .3 .4 .9 1.1 1.4]';
    X = [ones(size(x1)) x1 x2]
    y = [.17 .26 .28 .23 .27 .34]';
    w = [1 1 1 1 1 .1]';
    [b,se_b,mse] = lscov(X,y,w);
    plot(x1,y,'b', x2,y,'r');

Answer 1

[Marek Svoboda]

I found this StackExchange thread useful: using size of scatter points to weight line of best fit in matlab