daDoctoR/R/rep_lm.R

84 lines
2.1 KiB
R
Raw Normal View History

2018-10-03 10:32:10 +02:00
#' A repeated linear regression function
2018-10-02 21:07:43 +02:00
#'
2018-10-04 10:49:05 +02:00
#' For bivariate analyses, to determine which variables to include in adjusted model.
#' @param meas Effect meassure. Input as c() of columnnames, use dput().
#' @param vars variables in model. Input as c() of columnnames, use dput().
#' @param string variables to test. Input as c() of columnnames, use dput().
#' @param ci flag to get results as OR with 95% confidence interval.
#' @param data data frame to pull variables from.
2018-10-02 21:07:43 +02:00
#' @keywords linear regression
#' @export
#' @examples
2018-10-10 13:30:18 +02:00
#' rep_lm()
2018-10-03 10:32:10 +02:00
2018-10-04 10:49:05 +02:00
rep_lm<-function(meas,vars,string,ci=FALSE,data){
2018-10-03 10:32:10 +02:00
2018-10-02 21:07:43 +02:00
require(broom)
2018-10-03 10:32:10 +02:00
2018-10-02 21:07:43 +02:00
d<-data
2018-10-04 10:49:05 +02:00
x<-data.frame(d[,c(string)])
v<-data.frame(d[,c(vars)])
2018-10-10 13:30:18 +02:00
2018-10-04 10:49:05 +02:00
y<-d[,c(meas)]
2018-10-10 13:30:18 +02:00
dt<-cbind(y=y,v)
2018-10-04 10:49:05 +02:00
m1<-length(coef(lm(y~.,data = dt)))
2018-10-03 10:32:10 +02:00
2018-10-10 13:30:18 +02:00
names(v)<-c(vars)
if (is.factor(y)){stop("y should not be a factor!")}
2018-10-03 10:32:10 +02:00
2018-10-04 10:49:05 +02:00
if (ci==TRUE){
2018-10-03 10:32:10 +02:00
2018-10-04 21:06:22 +02:00
df<-data.frame(matrix(NA,ncol = 3))
2018-10-04 10:49:05 +02:00
names(df)<-c("pred","or_ci","pv")
2018-10-03 10:32:10 +02:00
2018-10-04 10:49:05 +02:00
for(i in 1:ncol(x)){
dat<-cbind(dt,x[,i])
m<-lm(y~.,data=dat)
2018-10-03 10:32:10 +02:00
2018-10-10 13:30:18 +02:00
ci<-suppressMessages(confint(m))
l<-round(ci[-c(1:m1),1],2)
u<-round(ci[-c(1:m1),2],2)
2018-10-04 10:49:05 +02:00
or<-round(coef(m)[-c(1:m1)],2)
or_ci<-paste0(or," (",l," to ",u,")")
pv<-round(tidy(m)$p.value[-c(1:m1)],3)
x1<-x[,i]
2018-10-03 10:32:10 +02:00
2018-10-04 10:49:05 +02:00
if (is.factor(x1)){
pred<-paste(names(x)[i],levels(x1)[-1],sep = "_")}
2018-10-03 10:32:10 +02:00
2018-10-04 10:49:05 +02:00
else {pred<-names(x)[i]}
2018-10-03 10:32:10 +02:00
2018-10-04 10:49:05 +02:00
df<-rbind(df,cbind(pred,or_ci,pv))}}
2018-10-03 10:32:10 +02:00
2018-10-10 13:30:18 +02:00
else {
2018-10-03 10:32:10 +02:00
2018-10-04 21:06:22 +02:00
df<-data.frame(matrix(NA,ncol = 3))
2018-10-04 10:49:05 +02:00
names(df)<-c("pred","b","pv")
2018-10-03 10:32:10 +02:00
2018-10-04 10:49:05 +02:00
for(i in 1:ncol(x)){
dat<-cbind(dt,x[,i])
2018-10-03 10:32:10 +02:00
2018-10-04 10:49:05 +02:00
m<-lm(y~.,data=dat)
b<-round(coef(m)[-c(1:m1)],3)
pv<-round(tidy(m)$p.value[-c(1:m1)],3)
x1<-x[,i]
2018-10-03 10:32:10 +02:00
2018-10-04 10:49:05 +02:00
if (is.factor(x1)){
pred<-paste(names(x)[i],levels(x1)[-1],sep = "_")
}
else {pred<-names(x)[i]}
df<-rbind(df,cbind(pred,b,pv))
}}
2018-10-03 10:32:10 +02:00
2018-10-04 10:49:05 +02:00
pa<-as.numeric(df[,3])
2018-10-10 13:30:18 +02:00
t <- ifelse(pa<=0.1,"include","drop")
2018-10-04 10:49:05 +02:00
pa<-ifelse(pa<0.001,"<0.001",pa)
pa <- ifelse(pa<=0.05|pa=="<0.001",paste0("*",pa),
ifelse(pa>0.05&pa<=0.1,paste0(".",pa),pa))
2018-10-03 10:32:10 +02:00
2018-10-04 10:49:05 +02:00
r<-data.frame(df[,1:2],pa,t)[-1,]
2018-10-03 10:32:10 +02:00
2018-10-04 10:49:05 +02:00
return(r)
2018-10-02 21:07:43 +02:00
}