sbh.control: Parameters Control Function

In jedazard/PRIMsrc: PRIM Survival Regression Classification

Parameters Control Function

Description

End-user function to set ancillary parameters of main end-user function sbh for fitting a Survival Bump Hunting (SBH) model. It is used to set some variable screening parameters, optional formats and outputs of sbh, as well as internally to tune the scatterplot smoother used for finding cross-validated model selection/tuning profile extremum.

Usage

  sbh.control(vscons = 0.5, 
              decimals = 2, 
              onese = FALSE, 
              probval = NULL, 
              timeval = NULL, 
              lag = 2, 
              span = 0.10, 
              degree = 2)

Arguments

vscons	numeric scalar in [1/K, 1], specifying the conservativeness of the variable screening (pre-selection) procedure, where 1/K is the least conservative and 1 is the most. Defaults to 0.5.
decimals	Positive integer of the number of user-specified significant decimals to output results. Defaults to 2.
onese	logical scalar. Flag for using the 1-standard error rule instead of extremum value of the cross-validation criterion when tuning/optimizing model parameters. Defaults to FALSE.
probval	numeric scalar in [0, 1] of the survival probability at which we want to get the endpoint box survival time. Defaults to NULL (i.e. maximal survival probability value is used).
timeval	numeric scalar of the survival time at which we want to get the endpoint box survival probability. Defaults to NULL (i.e. maximal survival time value is used).
lag	Positive integer indicating which lag to use in the lagged and iterated difference function. Defaults to 2.
span	numeric scalar in [0, 1], specifying the degree of smoothing in the internal stats::loess function. Defaults to 0.10. If span is too small with respect to the number of peeling steps (Adjusted maximum peeling length), choose a larger value such that floor(number of peeling steps * span) > 0.
degree	Positive integer indicating the degree of the polynomials (normally 1 or 2) to be used in the internal stats::loess function. Here, degree 0 is not also allowed unlike in stats::loess). Defaults to 2.

Details

Example of vscons values for pre-selection are as follows:

• '1.0' represents a presence in all the folds (unanimity vote)

• '0.5' represents a presence in at least half of the folds (majority vote)

• '1/K' represents a presence in at least one of the folds (minority vote)

Although any value in the interval [1/K,1] is accepted, we recommand using the interval [1/K, 1/2] to avoid excessive conservativeness. Final variable usage (selection) is done at the time of fitting the Survival Bump Hunting (SBH) model itself using our PRSP algorithm on previously screened variables by collecting those variables that have the maximum occurrence frequency in each peeling step over cross-validation folds and replicates.

Value

A list of 8 components.

Acknowledgments

This work made use of the High Performance Computing Resource in the Core Facility for Advanced Research Computing at Case Western Reserve University. This project was partially funded by the National Institutes of Health NIH - National Cancer Institute (R01-CA160593) to J-E. Dazard and J.S. Rao.

Note

End-user function to be used with sbh.

Author(s)

• "Jean-Eudes Dazard, Ph.D." [e-mail]

• "Michael Choe, M.D." [e-mail]

• "Michael LeBlanc, Ph.D." [e-mail]

• "Alberto Santana, MBA." [e-mail]

• "J. Sunil Rao, Ph.D." [e-mail]

Maintainer: "Jean-Eudes Dazard, Ph.D." [e-mail]

References

• Dazard J-E. and Rao J.S. (2021a). "Variable Selection Strategies for High-Dimensional Recursive Peeling-Based Survival Bump Hunting Models." (in prep).

• Dazard J-E. and Rao J.S. (2021b). "Group Bump Hunting by Recursive Peeling-Based Methods: Application to Survival/Risk Predictive Models." (in prep).

• Dazard J-E., Choe M., Pawitan Y., and Rao J.S. (2021c). "Identification and Characterization of Informative Prognostic Subgroups by Survival Bump Hunting." (in prep).

• Rao J.S., Huilin Y., and Dazard J-E. (2020). "Disparity Subtyping: Bringing Precision Medicine Closer to Disparity Science." Cancer Epidemiology Biomarkers & Prevention, 29(6 Suppl):C018.

• Yi C. and Huang J. (2017). "Semismooth Newton Coordinate Descent Algorithm for Elastic-Net Penalized Huber Loss Regression and Quantile Regression." J. Comp Graph. Statistics, 26(3):547-557.

• Dazard J-E., Choe M., LeBlanc M., and Rao J.S. (2016). "Cross-validation and Peeling Strategies for Survival Bump Hunting using Recursive Peeling Methods." Statistical Analysis and Data Mining, 9(1):12-42.

• Dazard J-E., Choe M., LeBlanc M., and Rao J.S. (2015). "R package PRIMsrc: Bump Hunting by Patient Rule Induction Method for Survival, Regression and Classification." In JSM Proceedings, Statistical Programmers and Analysts Section. Seattle, WA, USA. American Statistical Association IMS - JSM, p. 650-664.

• Dazard J-E., Choe M., LeBlanc M., and Rao J.S. (2014). "Cross-Validation of Survival Bump Hunting by Recursive Peeling Methods." In JSM Proceedings, Survival Methods for Risk Estimation/Prediction Section. Boston, MA, USA. American Statistical Association IMS - JSM, p. 3366-3380.

See Also

• sbh

• diff (R package base)

• loess (R package stats)

jedazard/PRIMsrc documentation built on July 16, 2022, 10:56 p.m.