plotEffects: Plotting time-varying effects from a risk regression model. In riskRegression: Risk Regression Models and Prediction Scores for Survival Analysis with Competing Risks

Description

Plot time-varying effects from a risk regression model.

Usage

 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18``` ```plotEffects( x, formula, level, ref.line = TRUE, conf.int = 0.95, xlim, ylim, xlab = "Time", ylab = "Cumulative coefficient", col, lty, lwd, add = FALSE, legend, axes = TRUE, ... ) ```

Arguments

 `x` Fitted object obtained with one of `ARR`, `LRR`, `riskRegression`. `formula` A formula to specify the variable(s) whose regression coefficients should be plotted. `level` For categorical variables the level (group) whose contrast to the reference level (group) should be plotted. `ref.line` Logical. If `TRUE` then add a horizontal line at zero. `conf.int` Logical. If `TRUE` then add confidence limits. Can be controlled using smart arguments. See examples `xlim` See `plot` `ylim` See `plot` `xlab` See `plot` `ylab` See `plot` `col` A vector of colors for the regression coefficients. `lty` A vector of line types for the regression coefficients. `lwd` A vector of line thicknesses for the regression coefficients. `add` Logical. If `TRUE` then add lines to an existing plot. `legend` Logical. If `TRUE` then add a legend. Can be controlled using smart arguments. See examples. `axes` Logical. If `FALSE` then do not draw axes. `...` Used for transclusion of smart arguments for `plot`, `axis`. See function `SmartControl` from prodlim.

Author(s)

Thomas H. Scheike ts@biostat.ku.dk

Thomas A. Gerds tag@biostat.ku.dk

Examples

 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31``` ```library(survival) library(prodlim) data(Melanoma) fit.tarr <- ARR(Hist(time,status)~strata(sex), data=Melanoma, cause=1) plotEffects(fit.tarr) fit.tarr <- ARR(Hist(time,status)~strata(sex)+strata(invasion), data=Melanoma, cause=1, times=seq(800,3000,20)) plotEffects(fit.tarr,formula=~sex) plotEffects(fit.tarr,formula=~invasion) plotEffects(fit.tarr, formula=~invasion, level="invasionlevel.1") ## legend arguments are transcluded: plotEffects(fit.tarr, formula=~invasion, legend.bty="b", legend.cex=1) ## and other smart arguments too: plotEffects(fit.tarr, formula=~invasion, legend.bty="b", axis2.las=2, legend.cex=1) ```

riskRegression documentation built on Jan. 13, 2021, 11:12 a.m.