The Student Room Group

Mathematical Modeling,

Hi, im doing a project in which we must find an equation that best fits the data we were given.

When I applied a polynomial function where the highest degree was 3, the Correlation Coefficient was equal to .9991. However, when I applied a polynomial function where the highest degree was 5, the correlation coefficient was equal to .9996.

Is there a reason why a function with a higher degree models the data better than one of a lower degree?
nk9230
Hi, im doing a project in which we must find an equation that best fits the data we were given.

When I applied a polynomial function where the highest degree was 3, the Correlation Coefficient was equal to .9991. However, when I applied a polynomial function where the highest degree was 5, the correlation coefficient was equal to .9996.

Is there a reason why a function with a higher degree models the data better than one of a lower degree?



Which data is it?
Reply 2
motoroller
Which data is it?

*EDIT*
It is Body Mass Index for Females from a particular year in the United States.

I first tried a Cubic, and then a Quintic.
nk9230
cubic
Trends probably can't be analysed mathematically. What reason would there be for the BMI to follow a cubic or quintic? In some physical situation like fluid flow, there might be an underlying cubic flow rate, but I don't think there is significance in data like this. All you can do is comment on the patterns observed from the data.
Reply 4
motoroller
Trends probably can't be analysed mathematically. What reason would there be for the BMI to follow a cubic or quintic? In some physical situation like fluid flow, there might be an underlying cubic flow rate, but I don't think there is significance in data like this. All you can do is comment on the patterns observed from the data.


Well, there has to be a funtion to model the data that was given to us, Its an IB Maths IA type II. I used a cubic based on the shape/ trend the data followed when plotted as a scatter plot.

Latest