coxSW {Coxmos}

coxSW

Description

The coxSW function conducts a stepwise Cox regression analysis on survival data, leveraging the capabilities of the My.stepwise R package. The primary objective of this function is to identify the most significant predictors for survival data by iteratively adding or removing predictors based on their statistical significance in the model. The resulting model is of class "Coxmos" with an attribute model labeled as "coxSW".

Usage

coxSW(
  X,
  Y,
  max.variables = 20,
  BACKWARDS = TRUE,
  alpha_ENT = 0.1,
  alpha_OUT = 0.15,
  toKeep.sw = NULL,
  initialModel = NULL,
  x.center = TRUE,
  x.scale = FALSE,
  remove_near_zero_variance = TRUE,
  remove_zero_variance = FALSE,
  toKeep.zv = NULL,
  remove_non_significant = FALSE,
  alpha = 0.05,
  MIN_EPV = 5,
  returnData = TRUE,
  verbose = FALSE
)

Arguments

X

Numeric matrix or data.frame. Explanatory variables. Qualitative variables must be transform into binary variables.

Y

Numeric matrix or data.frame. Response variables. Object must have two columns named as "time" and "event". For event column, accepted values are: 0/1 or FALSE/TRUE for censored and event observations.

max.variables

Numeric. Maximum number of variables you want to keep in the cox model. If MIN_EPV is not meet, the value will be change automatically (default: 20).

BACKWARDS

Logical. If BACKWARDS = TRUE, backward strategy is performed (default: TRUE).

alpha_ENT

Numeric. Maximum P-Value for a variable to enter the model (default: 0.10).

alpha_OUT

Numeric. Minimum P-Value for a variable to leave the model (default: 0.15).

toKeep.sw

Character vector. Name of variables in X to not be deleted by Step-wise selection (default: NULL).

initialModel

Character vector. Name of variables in X to include in the initial model (default: NULL).

x.center

Logical. If x.center = TRUE, X matrix is centered to zero means (default: TRUE).

x.scale

Logical. If x.scale = TRUE, X matrix is scaled to unit variances (default: FALSE).

remove_near_zero_variance

Logical. If remove_near_zero_variance = TRUE, near zero variance variables will be removed (default: TRUE).

remove_zero_variance

Logical. If remove_zero_variance = TRUE, zero variance variables will be removed (default: TRUE).

toKeep.zv

Character vector. Name of variables in X to not be deleted by (near) zero variance filtering (default: NULL).

remove_non_significant

Logical. If remove_non_significant = TRUE, non-significant variables/components in final cox model will be removed until all variables are significant by forward selection (default: FALSE).

alpha

Numeric. Numerical values are regarded as significant if they fall below the threshold (default: 0.05).

MIN_EPV

Numeric. Minimum number of Events Per Variable (EPV) you want reach for the final cox model. Used to restrict the number of variables/components can be computed in final cox models. If the minimum is not meet, the model cannot be computed (default: 5).

returnData

Logical. Return original and normalized X and Y matrices (default: TRUE).

verbose

Logical. If verbose = TRUE, extra messages could be displayed (default: FALSE).

Details

The coxSW function employs a stepwise regression technique tailored for survival data. This method is particularly beneficial when dealing with a plethora of predictors, and there's a necessity to distill the model to its most impactful variables. The stepwise procedure can be configured to operate in forward, backward, or a hybrid mode, contingent on the parameters specified by the user.

During the iterative process, variables are evaluated for inclusion or exclusion based on predefined significance levels (alpha_ENT for entry and alpha_OUT for removal). This ensures that the model retains only those predictors that meet the significance criteria, thereby enhancing the model's interpretability and predictive power.

Additionally, the function offers several preprocessing options, such as centering and scaling of the predictor matrix, removal of variables with near-zero or zero variance, and the ability to enforce the inclusion of specific variables in the model. These preprocessing steps are crucial for ensuring the robustness and stability of the resulting Cox regression model.

It's worth noting that the function is equipped to handle both numeric and binary categorical predictors. However, it's imperative that categorical variables are appropriately transformed into binary format before analysis. The outcome or response variable should comprise two columns: "time" representing the survival time and "event" indicating the occurrence of the event of interest.

Value

Instance of class "Coxmos" and model "coxSW". The class contains the following elements:

X: List of normalized X data information.

• (data): normalized X matrix

• (x.mean): mean values for X matrix

• (x.sd): standard deviation for X matrix

Y: List of normalized Y data information.

• (data): normalized Y matrix

• (y.mean): mean values for Y matrix

• (y.sd): standard deviation for Y matrix

survival_model: List of survival model information

• fit: coxph object.

• AIC: AIC of cox model.

• BIC: BIC of cox model.

• lp: linear predictors for train data.

• coef: Coefficients for cox model.

• YChapeau: Y Chapeau residuals.

• Yresidus: Y residuals.

call: call function

X_input: X input matrix

Y_input: Y input matrix

nsv: Variables removed by remove_non_significant if any.

nzv: Variables removed by remove_near_zero_variance or remove_zero_variance.

nz_coeffvar: Variables removed by coefficient variation near zero.

removed_variables_correlation: Variables removed by being high correlated with other variables.

class: Model class.

time: time consumed for running the cox analysis.

Author(s)

Pedro Salguero Garcia. Maintainer: [e-mail]

References

Efroymson MA (1960). “Multiple Regression Analysis.” Mathematical Methods for Digital Computers. Company ISC (2017). “My.stepwise: Stepwise Variable Selection Procedures for Regression Analysis.” https://cran.r-project.org/package=My.stepwise.

Examples

data("X_proteomic")
data("Y_proteomic")
X <- X_proteomic[,1:10]
Y <- Y_proteomic
coxSW(X, Y, x.center = TRUE, x.scale = TRUE)