predict.BLMCP

Fitted <code>"BLMCP"</code> object.

object

Matrix of new values for <code>E</code> at which predictions are to be
made.

newE

Matrix of new values for <code>G</code> at which predictions are to be
made.

newG

Not used. Other arguments to predict.

&#8230;

This function makes predictions from a BLMCP
model, using the stored <code>"BLMCP"</code> object.

Description: For the risk, progression, and response to treatment of many complex diseases, it has been increasingly recognized that gene-environment interactions play important roles beyond the main genetic and environmental effects. In practical interaction analyses, outliers in response variables and covariates are not uncommon. In addition, missingness in environmental factors is routinely encountered in epidemiological studies. The developed package consists of five robust approaches to address the outliers problems, among which two approaches can also accommodate missingness in environmental factors. Both continuous and right censored responses are considered. The proposed approaches are based on penalization and sparse boosting techniques for identifying important interactions, which are realized using efficient algorithms. Beyond the gene-environment analysis, the developed package can also be adopted to conduct analysis on interactions between other types of low-dimensional and high-dimensional data. (Mengyun Wu et al (2017), <doi:10.1080/00949655.2018.1523411>; Mengyun Wu et al (2017), <doi:10.1002/gepi.22055>; Yaqing Xu et al (2018), <doi:10.1080/00949655.2018.1523411>; Yaqing Xu et al (2019), <doi:10.1016/j.ygeno.2018.07.006>; Mengyun Wu et al (2021), <doi:10.1093/bioinformatics/btab318>).

predict.BLMCP: Make predictions from a "BLMCP" object

Description

Usage

Arguments

Value

References

See Also