https://github.com/mayer79/r-confidence-intervals-r-squared

Confidence intervals on R-squared
https://github.com/mayer79/r-confidence-intervals-r-squared

Last synced: 15 days ago
JSON representation

Confidence intervals on R-squared

• Host: GitHub
• URL: https://github.com/mayer79/r-confidence-intervals-r-squared
• Owner: mayer79
• Created: 2017-03-27T08:38:40.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2017-03-27T08:46:28.000Z (over 7 years ago)
• Last Synced: 2024-10-04T12:56:53.251Z (3 months ago)
• Language: R
• Size: 1.95 KB
• Stars: 2
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# R-confidence-intervals-R-squared

R function that takes an `lm' object and returns and approximate one- or two-sided confidence interval for population R-squared.

It helps to judge how strong the linear relation between regressors and response truely is with given (approximate) certainty. A positive lower bound of a one-sided confidence interval means the corresponding global F-test provides a significant result on the corresponding level.

## Example

```

  library(MBESS)

  fit <- lm(Sepal.Length ~ Petal.Width, data = iris) # R-squared: 0.669

  confintR2(fit, alternative = "greater")            # Lower 95% limit for true R-squared: 0.5982115

```