alexpkeil1

# using BKMRhat with multiple imputation

library(bkmrhat)
library(mice)
library(bkmr)
library(future)

# create simulated dataset for testing
dat <- bkmr::SimData(n = 50, M = 4)

alexpkeil1

library(qgcomp)
library(survival)
library(ggplot2)
set.seed(50)
N=200
dat <- data.frame(time=(tmg <- pmin(.1,rweibull(N, 10, 0.1))), 
                  d=1.0*(tmg<0.1), x1=runif(N), x2=runif(N), z=runif(N))
expnms=paste0("x", 1:2)
f = survival::Surv(time, d)~x1 + x2 + I(x2*x1)
## Not run: 

alexpkeil1

# simple example of using Bayes to penalize estimates in quantile g-computation

library(qgcomp)
library(MASS)
n= 50
Sigma <- matrix(c(10,3,3,2),2,2)
Sigma

set.seed(1232)
X = MASS::mvrnorm(n, mu=rep(0,2), Sigma)

alexpkeil1

# simple application of qgcomp with time-varying exposures
# the simulated model is for the discrete hazard, given recent exposure and recent values of confounders
# this method can be used for alternative scenarios (e.g. cumulative exposure), which corresponds to 
# different parametric assumptions

library(qgcomp)
library(survival)

# create simple data set with time-varying exposures
makedata <- function(N=100){

alexpkeil1

# scoping issue when using in-formula transformations
# example: spline basis functions
# problem: when including functions like `rms(x)` in a formula to put a spline
#   function on 'x', plotting a qgcomp.cox.boot object yields an error that 'x' can't be found
# solution: explicitly create the spline basis functions as variables in a data.frame. Example follows.

library(qgcomp)
library(rms)
library(survival)

alexpkeil1

# qgcomp with a gee like approach using bootstrapping or an estimating equation based approach - useful for clustered or longitudinal data when the interest is in effect of x -> y, pooled over multiple time points

library(qgcomp)
library(qgcompint)
set.seed(50)

####### simulate some clustered data -----
# linear model, binary modifier
# simulate cluster specific exposures and outcome means by just treating these like independent observations
dat <- qgcompint::simdata_quantized_emm(outcometype = "continuous", 

alexpkeil1

library(qgcomp)

msm.fit.noint <- function(f,
                          qdata,
                          intvals,
                          expnms,
                          rr=TRUE,
                          main=TRUE,
                          degree=1,
                          id=NULL,

alexpkeil1

library(qgcomp)
library(mvtnorm)

Niter = 1000
seedvals = round(runif(Niter)*.Machine$integer.max)


truemod <- function(X,beta){
  X %*% beta
}

alexpkeil1

# using qgcomp.cox.noboot in a nested case-control analysis as a way to fit a conditional logit qgcomp model
library(qgcomp)
library(survival)

# read in data from a nested case control study
data(infert, package = "datasets")
head(infert)


### 1) check that it works, part A

alexpkeil1

#!/usr/local/env R
# a way to base the quantiles off of weighted data

library(qgcomp)
library(DescTools) # need this to calculate weighted quantiles
data(metals)

# new function to calculate weighted quantiles
quantize.wtd <- function (data, expnms, q = 4, breaks = NULL, weights=NULL) 
{