https://github.com/htpusa/stabilityselection

https://github.com/htpusa/stabilityselection

Last synced: 20 days ago
JSON representation

• Host: GitHub
• URL: https://github.com/htpusa/stabilityselection
• Owner: htpusa
• Created: 2023-12-14T14:02:58.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2023-12-14T14:15:29.000Z (about 1 year ago)
• Last Synced: 2023-12-14T15:31:38.117Z (about 1 year ago)
• Language: MATLAB
• Size: 2.93 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# stabilitySelection

Perform stability selection using the MATLAB function lassoglm. The function `stabilityPaths` estimates the probability of a variable to be chosen in a sparse model by repeatedly subsampling the data, calculating the reguralisation path for each subsample, and counting how frequently the variable was chosen at each point in the path.

The resulting *stability paths* can be used to select or rank variables based on, for example, the maximum selection probability they achieved.

For more on the method, see Meinshausen & Bühlmann, 2010, and Shah & Samworth, 2013.

Meinshausen, Nicolai, and Peter Bühlmann. "Stability selection." 

   Journal of the Royal Statistical Society Series B: Statistical 

   Methodology 72.4 (2010): 417-473.

   

Shah, Rajen D., and Richard J. Samworth. "Variable selection with error

   control: another look at stability selection." Journal of the Royal

   Statistical Society Series B: Statistical Methodology 75.1 (2013): 55-80.

# Example

```

a = [1:-0.1:0.1 zeros(1,40)];

Y = randn(100,1);

X = normrnd(Y*a,0.2);

Y = Y>0;

[selProb maxProb numSel Lambda] = stabilityPaths(X,Y);

plotStabilityPaths(selProb,1:10);

```