Welcome to our community

Be a part of something great, join today!

Require assistance with possible multiple regression analysis

Dants

New member
Apr 9, 2019
7
I am interested in determining more efficient ways of determining individuals' body fat percentage. To do this, I measure the circumference of a number of segments (10 of them) of the body and determine the person's percentage body fat through underwater weighing. I have done this for 252 total participants, and I have recorded the age for the participants.

My goal is to facilitate the simple estimation of percentage body fat. I wish to generate a predictive equation the best estimate of percentage body fat. Am I correct to assume that I should be using multiple regression or simple linear regression?

Thanks,

Dants
 

Ackbach

Indicium Physicus
Staff member
Jan 26, 2012
4,184
In a machine learning environment, the terms are essentially interchangeable, and it doesn't matter. The underlying model is the same either way. In fact, any kind of regression in which the coefficients show up linearly is still linear regression, even if the shape of the line or curve being fit is not straight. It's how the coefficients show up that determine if the model is a linear regression model, not the shape of the line or curve.