So the math should be similar: you should be adding a couple intercept translation figures multiplying by a scaling factor. The major thing that would change is how you would find those scaling ...