My Math Forum What Type Of Regression Should I Do With This Data?

 Probability and Statistics Basic Probability and Statistics Math Forum

 April 5th, 2017, 06:43 AM #1 Senior Member   Joined: Oct 2013 From: New York, USA Posts: 541 Thanks: 77 What Type Of Regression Should I Do With This Data? I have radio market populations. The difference in population between consecutive market numbers is bigger for bigger markets. Since the data is not linear, what type of regression should I do? My calculator can do these types of nonlinear regression: Quadratic Cubic Quartic Natural logarithm Exponential Power Logistic Sine
 April 5th, 2017, 09:15 AM #2 Member   Joined: Jan 2017 From: California Posts: 79 Thanks: 8 Plot the data on a scattergraph look at the shape and go from there
 April 5th, 2017, 03:26 PM #3 Senior Member   Joined: Oct 2013 From: New York, USA Posts: 541 Thanks: 77 There are 274 numbers, and the difference between them is big enough that it wouldn't look nice on a scatterplot unless the y scale went up exponentially. One way to think about it is the salary of 274 employees, with the richest few bringing the mean way above the median.
April 5th, 2017, 04:55 PM   #4
Math Team

Joined: Jul 2011
From: Texas

Posts: 2,611
Thanks: 1297

Quote:
 ... unless the y scale went up exponentially
if the data set models an exponential function, then the data plot of x vs. log(y) would look linear

April 6th, 2017, 03:38 PM   #5
Senior Member

Joined: Oct 2013
From: New York, USA

Posts: 541
Thanks: 77

Quote:
 Originally Posted by skeeter if the data set models an exponential function, then the data plot of x vs. log(y) would look linear
I don't know if my calculator can use a logarithmic scale.

Is there a way of doing different types of regression, finding a value of how well each model fits the data, and therefore a way to tell which type of regression fit the best?

 Tags data, regression, type

 Thread Tools Display Modes Linear Mode

 Similar Threads Thread Thread Starter Forum Replies Last Post Davuit Advanced Statistics 0 May 9th, 2013 09:24 AM sydfremmer Advanced Statistics 0 January 29th, 2013 05:52 AM sydfremmer Advanced Statistics 0 December 22nd, 2012 08:26 AM lumpa Real Analysis 0 October 19th, 2012 08:07 AM nyalex Advanced Statistics 1 August 24th, 2011 09:24 AM

 Contact - Home - Forums - Cryptocurrency Forum - Top