A function giving prediction information, from a nearest shrunken centroid fit

1 2 3 | ```
pamr.predict(fit, newx, threshold,
type = c("class", "posterior", "centroid", "nonzero"),
prior = fit$prior, threshold.scale = fit$threshold.scale)
``` |

`fit` |
The result of a call to pamr.train |

`newx` |
Matrix of features at which predictions are to be made |

`threshold` |
The desired threshold value |

`type` |
Type of prediction desired: class predictions, posterior probabilities, (unshrunken) class centroids, vector of genes surviving the threshold |

`prior` |
Prior probabilities for each class. Default is that specified in "fit" |

`threshold.scale` |
Additional scaling factors to be applied to the thresholds. Vector of length equal to the number of classes. Default is that specified in "fit". |

`pamr.predict`

Give a cross-tabulation of true versus
predicted classes for the fit returned by pamr.train or pamr.cv,
at the specified threshold

Trevor Hastie, Robert Tibshirani, Balasubramanian Narasimhan, and Gilbert Chu

1 2 3 4 5 6 7 8 |

Questions? Problems? Suggestions? Tweet to @rdrrHQ or email at ian@mutexlabs.com.

All documentation is copyright its authors; we didn't write any of that.