# Thread: Converting units/scale in a linear regression model

1. ## Converting units/scale in a linear regression model

So this is a very simple question, but it's been years since I did regression and I can't quite figure it out
So basically, I'm given a regression function for predicted weight (w = -99.41 + 3.94 x height) which is in terms of inches and pounds. The R^2 = 0.81 and SER = 10.2
The question is, how can I convert that regression function into kilograms and centimeters? I've tried multiplying by the conversion factors, and for some reason I just can't seem to get it Also, what would happen to the SER? I know the R^2 should stay the same, because it's unit-less and such. Anyone able to help me out here? It would be greatly appreciated if you could point me in the right direction! Thanks!!

2. Originally Posted by mistykz
So this is a very simple question, but it's been years since I did regression and I can't quite figure it out
So basically, I'm given a regression function for predicted weight (w = -99.41 + 3.94 x height) which is in terms of inches and pounds. The R^2 = 0.81 and SER = 10.2
The question is, how can I convert that regression function into kilograms and centimeters? I've tried multiplying by the conversion factors, and for some reason I just can't seem to get it Also, what would happen to the SER? I know the R^2 should stay the same, because it's unit-less and such. Anyone able to help me out here? It would be greatly appreciated if you could point me in the right direction! Thanks!!

$\displaystyle w_{kg}=0.4536 \times w_{lb} = 0.4536 (-99.41 + 3.94 \times h_{inches})=-45.09+1.79\times h_{inches}$

and

$\displaystyle h_{inches}=h_{cm}/2.54$

so:

$\displaystyle w_{kg}=-45.09+0.7036\times h_{cm}$

The Standard Error of the Regression has units of mass so you multiply it by $\displaystyle 0.4536$ to convert it to $\displaystyle kg$.

CB

3. Thank you so much!