Description Usage Arguments Value References
Wrapper function to run STRIDE estimators that computes the distribution function for the mixture data based on an expectation-maximization (EM) algorithm that uses the pool adjacent violators algorithm (PAVA) from isotone regression to yield a non-negative and monotone estimator. This estimator ignores covariates.
1 | stride.empava(n, q, x, delta, timeval, p, tol.level = 1e-05, count.max = 100)
|
n |
sample size, must be at least 1. |
q |
a numeric matrix of size |
x |
a numeric vector of length |
delta |
a numeric vector of length |
timeval |
numeric value at which the distribution function is evaluated. |
p |
number of populations, must be at least 2. |
tol.level |
a numeric that denotes the tolerance level of the average L1-difference of estimates of the distribution function between EM iterations. Default is 0.00001. |
count.max |
a numeric that denotes the number of EM iterates to obtain. Default is 100. |
a list containing
Fest: numeric array of dimension p
by length(timeval)
containing EM-PAVA estimates of the distribution function evaluated at timeval
.
Fest.all: numeric array of dimension p
by length(timeval)+length(x)
containing EM-PAVA estimmates of the distribution function evaluated at all timeval
and all x
values.
Garcia, T.P., Marder, K. and Wang, Y. (2017). Statistical modeling of Huntington disease onset. In Handbook of Clinical Neurology, vol 144, 3rd Series, editors Andrew Feigin and Karen E. Anderson.
Qing, J., Garcia, T.P., Ma, Y., Tang, M.X., Marder, K., and Wang, Y. (2014). Combining isotonic regression and EM algorithm to predict genetic risk under monotonicity constraint. Annals of Applied Statistics, 8(2), 1182-1208.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.