hd.study: Analysis example using the simulated HD dataset

In unkyunglee/HDChangePoint: Estimating disease onset from change points of markers measured with error

Description

Analysis example using the simulated HD dataset

Usage

hd.study(
  simu.data = simu.dat,
  subid = "SUBJID",
  event = "event",
  tms = "TOTAL_MOTOR_SCORE",
  cag = "CAG",
  age = "AGE",
  gender = "gender",
  trans.age = "logAGE",
  m = 30,
  num.interp = 30,
  n = 80,
  newl = 30,
  mean.diff = 1,
  tolerance = 0.005,
  itermax = 20,
  iter = 0,
  boot.ci = TRUE
)

Arguments

simu.data	a simulated data set, see DATA.R for details.
subid	column name for the group id from your dataset.
event	column name for the event of interest from your dataset.
tms	column name for the total motor score from your dataset.
cag	column name for the cag repeats from your dataset.
age	column name for the age (time) from your dataset.
gender	column name for the gender from your dataset.
trans.age	log transformed age (time) from your dataset.
m	number of time points at which predictors are required for the longitudinal responses in the parametric NLME procedure
num.interp	number of pseudo-data points, which are unifromly generated within the observed data points.
n	number of sample size.
newl	a length of time points (log transformed ages) at which predictors are required for each individual longitudinal trajectory.
mean.diff	an initial tolerence for the convergence criterion. Default is 1.
tolerance	a parameter for the tolerence for the convergence criterion. Default is 0.009.
itermax	a parameter for the maximum number of iterations in the multi-stage nonparametric algorithm. Default is 20.
iter	initial number of iteration in the multi-stage nonparametric algorithm. The default is 0.
boot.ci	logical value: TRUE if the 95% bootstrap confidence intervals across all subjects are calcuated and plotted on the graph. Default is TRUE.

Value

a list of

• nonpara_summary_table4a 5 x 4 array of estimates, standard errors, t-values and p-values for the fixed effects of beta0, CAG repeates and gender in the inflection point model and scale parameters of the random errors in the inflection point and the longitudinal models.

• para_summary_table4a 7 x 4 array of estimates, standard errors, t-values and p-values for the fixed effects of theta1, theta2, beta0, CAG repeates and gender and scale parameters of the random errors in the inflection point and the longitudinal models. The parametric NLME assumes the logistic model. , which are similar to Table 4 and return Figure 1 in the paper (See the reference).

References

U.Lee, R.J.Carroll, K.Marder, Y.Wang, T.P.Garcia. (2019+). Estimating Disease Onset from Change Points of Markers Measured with Error.

Examples

library(HDChangePoint)
## a pseudo data set constructed by using the simulated data
data("PSEUDO_PREDICT_HD")
head(PSEUDO_PREDICT_HD)

## Specify the parameters to obtained the analysis results from the simulated dataset.
simu.dat<-PSEUDO_PREDICT_HD
subid="SUBJID";
event="event";
tms="TOTAL_MOTOR_SCORE";
cag="CAG";
age="AGE";
gender="gender";
trans.age="logAGE";
n=80;
m=45;
num.interp=45;
newl=45;
mean.diff=1;
tolerance=0.01;
itermax=20;
iter=0;

simu.analysis.results<-hd.study(simu.data=simu.dat, subid="SUBJID", event="event", 
                       tms="TOTAL_MOTOR_SCORE", cag="CAG", age="AGE", gender="gender", 
                       trans.age="logAGE",m=m, num.interp=num.interp, n=n, newl=newl,
                       mean.diff=mean.diff, tolerance=tolerance, itermax=itermax, iter=iter,
                       boot.ci=TRUE)

unkyunglee/HDChangePoint documentation built on Nov. 27, 2021, 2:57 p.m.