sssResult-class: Results from call to sss method - '"sssResult"'

In Sage-Bionetworks/sss: Shotgun Stochastic Search (SSS) from Hans, Dobra and West

Description

A generic result object that contains information about the model that was run as well as the results from running the sss method.

Objects from the Class

Objects are returned by default from the "sss" algorithm

Slots

standScore:

Object of class "numeric" - the standardized scores for the n best models

postMargProb:

Object of class "numeric" - posterior marginal probabilities of variables included in the n best models sorted in decending order

testPredictionSummary:

Object of class "numeric" - weighted average of predictions for each sample passed in the testing set based on standScore

model:

Object which extends on class "sssModel" depending on type of model fit

nBestFits

Object of class "list" - specific information about the n best model fits - see next section for specifics for each type of model.

nBestFits - sssBinaryModel

p:

Object of class "list" - number of predictors for this model (each list entry represents the i-th model)

score:

Object of class "list" - log posterior probability of this model (each list entry represents the i-th model)

indices:

Object of class "list" - the indices of the p variables in this model (each list entry represents the i-th model)

pmode:

Object of class "list" - posterior mode of the regression parameter vector beta including the intercept (each list entry represents the i-th model)

pvar:

Object of class "list" - estimated posterior variance matrix of beta in vectorized form including the intercept (each list entry represents the i-th model)

trainPrediction:

Object of class "list" - predictions on internal training set

testPrediction:

Object of class "list" - predictions on test set (if available) based on models fit by training set

nBestFits - sssLinearModel

p:

Object of class "list" - number of predictors for this model (each list entry represents the i-th model)

score:

Object of class "list" - log posterior probability of this model (each list entry represents the i-th model)

indices:

Object of class "list" - the indices of the p variables in this model (each list entry represents the i-th model)

pmean:

Object of class "list" - posterior mean of the regression parameter vector beta excluding the intercept (each list entry represents the i-th model)

pvar:

Object of class "list" - posterior variance matrix of beta in vectorized form excluding the intercept (each list entry represents the i-th model)

residsd:

Object of class "list" - residual SD estimate (each list entry represents the i-th model)

postdf:

Object of class "list" posterior degrees of freedom (each list entry represents the i-th model)

trainPrediction:

Object of class "list" - predictions on internal training set

testPrediction:

Object of class "list" - predictions on test set (if available) based on models fit by training set

nBestFits - sssSurvivalModel

p:

Object of class "list" - number of predictors for this model (each list entry represents the i-th model)

score:

Object of class "list" - log posterior probability of this model (each list entry represents the i-th model)

indices:

Object of class "list" - the indices of the p variables in this model (each list entry represents the i-th model)

pmeanalpha:

Object of class "list" - posterior mean of the Weibull index parameter in this model (each list entry represents the i-th model)

pmode:

Object of class "list" - posterior mode of the regression parameter vector beta including the intercept (each list entry represents the i-th model)

pvar:

Object of class "list" - estimated posterior variance matrix of (alpha, beta) including intercept in vectorized form (each list entry represents the i-th model)

trainPrediction:

Object of class "list" - predictions on internal training set

testPrediction:

Object of class "list" - predictions on test set (if available) based on models fit by training set

See Also

model classes

sssModel, sssBinaryModel, sssLinearModel, sssSurvivalModel

setup class

sssSetup

methods

sss

Examples

showClass("sssResult")

Sage-Bionetworks/sss documentation built on May 9, 2019, 12:14 p.m.