engleski

The best total least squares line in $R^3$

We consider the problem of determining the line in $mathbb R^3$ which does the best total least squares approximation of the given data $(p_i,{f r}_i),,i=1,ldots,m$, where ${f r}_i=(x_i,y_i,z_i)^TinR^3$ are some points, and $p_i>0$ are corresponding weights. By using elementary mathematics, we show how to determine such a best total least squares line.

total least squares; orthogonal distance regression

nije evidentirano