Hi Experts,

Please find attached an excel spreadsheet containing 2 charts and 4 data plots. What form of regression would most accurately model this data? At the moment I have a linear plot selected but it seems to me the data plots have a subtle s shape.
Who is Participating?
Those are actaully pretty good fits.

What is the data, and why do think there should be any fit at all?

I do see something in the data that looks vaguely sinusoidal.
You might try a Fourier Transform to see if that's the case.
What relationship makes the most sense for the kind of data you have?

You can produce a fit to random noise and allowing more subtle shapes can produce better fits to random noise, but they would not be any more meaningful.

Asking for an arbitrary model for arbitrary data makes no sense.

You need to start by asking what kind of model makes sense for what the data represents.
DColinAuthor Commented:
Hi d-glitch:,

"You might try a Fourier Transform to see if that's the case."

Can excel do this?
Free Tool: IP Lookup

Get more info about an IP address or domain name, such as organization, abuse contacts and geolocation.

One of a set of tools we are providing to everyone as a way of saying thank you for being a part of the community.

DColinAuthor Commented:
Hi ozo,

The X-axis represents ability, 1 best 20 not best. Y-axis represents earnings.

I have no idea what relationships I am looking for. I was hoping these data plots would tell me what the relationships are.
hi, I too thought the same as Dglitch, that there might be a sine wave in there, but there seems little logical explanation of why that would be, given your explanation of what the graphs represent.  

What your graphs show is that earnings and ability are positively correlated in a linear fashion.  Your X axis actually represents diminishing ability as you move along it, so the score 1 to 20 represents the amount of uselessness of the person
DColinAuthor Commented:
Hi deighton,

Would an explanation of the potential sine wave be that I have ranked the salespeople on the X-axis based on past performance. If I were to enumerate this rank it would not go 1,2,3 etc. but something like 100, 89,88,87,72,70,65 (best to worst). Do you think the non linearity of a ranking system is causing this?
I doubt if there is a sine wave effect at all.  Seeing the sine wave was a risk of looking at things too closely.  It looks pretty much linear.  Absolute linearity is very unlikely to occur, there will always be some sort of variation.
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.