The Student Room Group
Uni students - Complete our survey >>
Uni students - Complete our survey >>
Uni students - Complete our survey >>

Maths - Regression Question?

Hi Everyone,

For a maths project I am doing, I have modelled some data using a cosine function and I wanted to have a quantitative way to know how well the model fits the data points. Can I find an r^2 value for a cosine graph, because I have read that it can only be used for linear regression. However, when I did a sine regression on Desmos it gave me an r^2 value.

Thanks in advance
Reply 1
Avatar for Amefish
Amefish
20
Original post by Dotti22
Hi Everyone,

For a maths project I am doing, I have modelled some data using a cosine function and I wanted to have a quantitative way to know how well the model fits the data points. Can I find an r^2 value for a cosine graph, because I have read that it can only be used for linear regression. However, when I did a sine regression on Desmos it gave me an r^2 value.

Thanks in advance


Hi! I'm not very great at maths, but I'll move your question over to the maths section so you get replies from our maths geniuses :biggrin:
Reply 2
Avatar for Dotti22
Dotti22
OP
2
Original post by Amefish
Hi! I'm not very great at maths, but I'll move your question over to the maths section so you get replies from our maths geniuses :biggrin:


Thank you so much!
Original post by Dotti22
Hi Everyone,

For a maths project I am doing, I have modelled some data using a cosine function and I wanted to have a quantitative way to know how well the model fits the data points. Can I find an r^2 value for a cosine graph, because I have read that it can only be used for linear regression. However, when I did a sine regression on Desmos it gave me an r^2 value.

Thanks in advance


Yes, this will work fine - the "linear" in "linear regression" refers to linearity in the regression coefficients - the betas that you estimate, not in the functional forms of the covariates. You can fit linear combinations of any functions that you like using linear regression.

Quick Reply

Related discussions

Latest

Trending

Trending

Articles for you