Stats help - Linear Model fitting in R

Click For Summary
The discussion focuses on issues with a residual vs fitted values plot in R, where the variance shows a diamond shape instead of the expected constant variance and mean of zero. Attempts to resolve this through response variable transformations, including raising the variable to a power less than one and applying a log transformation, have not been successful. Participants inquire about the number of explanatory variables, the modeling approach, and whether the intercept is constrained to zero. They suggest that a response transformation may not be effective given the observed linear trend in the residual plot. The conversation emphasizes the need for further details, including the specific R code being used, to provide more targeted assistance.
aerot89
Messages
4
Reaction score
0
I have this residual vs fitted values plot in R, and it should show a scatter of points with constant variance and mean 0. The mean 0 part seems to hold, but there is quite a clear diamond shape for the variance...

I have tried raising the response variable to a power <1 and a log transformation, along with transforming all of the parameter types but all to no avail.

Any help would be greatly appreciated. Anyone have any tips as how to try and improve this...

Thanks

QpY3Z.png
 
Physics news on Phys.org
How many explanatory variables do you have? What are you trying to model? Are you restricting your intercept to a value of zero? From the clear linear trend in your residual plot, a response transformation will not help your situation. What is the R code that you are using?
 

Thread 'My basic understanding of set theory'

If there are an infinite number of natural numbers, and an infinite number of fractions in between any two natural numbers, and an infinite number of fractions in between any two of those fractions, and an infinite number of fractions in between any two of those fractions, and an infinite number of fractions in between any two of those fractions, and... then that must mean that there are not only infinite infinities, but an infinite number of those infinities. and an infinite number of those...
View full post »

Similar threads

Undergrad Linear Models vs Nonlinear Models
  • · Replies 22 ·
Replies
22
Views
4K
Undergrad Linear Model with independent categorical variable
  • · Replies 5 ·
Replies
5
Views
2K
Undergrad What is the expected value of Cov(x,y)2 in an independent X and Y scenario?
  • · Replies 17 ·
Replies
17
Views
3K
Undergrad Fitting two models in sequence with parameters in common
  • · Replies 12 ·
Replies
12
Views
2K
Undergrad What are the commonly used estimators in regression models?
  • · Replies 23 ·
Replies
23
Views
4K
Undergrad Nonlinear Least Squares or OLS for Nonlinear Models?
  • · Replies 7 ·
Replies
7
Views
2K
Undergrad Which method should I use for a linear fit of data with error on both z and y?
  • · Replies 8 ·
Replies
8
Views
2K
Undergrad About chi-squared and r-squared test for fitting data
  • · Replies 5 ·
Replies
5
Views
9K
Undergrad Linear fit on the difference of data points
  • · Replies 26 ·
Replies
26
Views
3K
Undergrad What Are the Key Differences Between OLS and GLM Models in Statistical Analysis?
  • · Replies 3 ·
Replies
3
Views
3K