Given a desired type I error rate and a stability path calculated with `stability.path`

the function selects a stable set of variables.

stabsel(x,error=0.05,type=c("pfer","pcer"),pi_thr=0.6)

## Arguments

x |
an object of class "stabpath" as returned by the function `stabpath` . |

error |
the desired type I error level w.r.t. to the chosen type I error rate. |

type |
The type I error rate used for controlling the number falsely selected variables. If `type="pfer"` the per-family error rate is controlled and `error` corresponds to the expected number of type I errors.
Selecting `type="pfer"` and `error` in the range of $0 > `error` < 1$ will control the family-wise error rate, i.e. the probability that at least one variable in the estimated stable set has been falsely selected.
If `type="pcer"` the per-comparison error rate is controlled and `error` corresponds to the expected number of type I errors divided by the number variables. |

pi_thr |
the threshold used for the stability selection, should be in the range of $0.5 > pi_thr < 1$. |

## Value

a list of four objects

stablea vector giving the positions of the estimated stable variables

lambdathe penalization parameter used for the stability selection

lposthe position of the penalization parameter in the regularization path

errorthe desired type I error level w.r.t. to the chosen type I error rate

typethe type I error rate

## Author

Martin Sill \
m.sill@dkfz.de

## References

Meinshausen N. and B\"uhlmann P. (2010), *Stability Selection, Journal of the Royal Statistical Society: Series B (Statistical Methodology) Volume 72, Issue 4, pages 417--473.*

Sill M., Hielscher T., Becker N. and Zucknick M. (2014), *c060: Extended Inference with Lasso and Elastic-Net Regularized Cox and Generalized Linear Models, Journal of Statistical Software, Volume 62(5), pages 1--22.*
https://www.jstatsoft.org/v062/i05/

## See also

## Examples