stride.nocovariates: STRIDE estimators without covariates
In tpgarcia/stride: STRIDE: Robust powerful mixture models

Description Usage Arguments Value References

Wrapper function to run STRIDE estimators that ignore covariates.

stride.nocovariates(
  n,
  q,
  x,
  delta,
  timeval,
  qvs,
  p,
  m,
  r,
  boot,
  bootvar,
  useOLS,
  useWLS,
  useEFF,
  useNPMLEs,
  useEMPAVA
)

`n`	sample size, must be at least 1.
`q`	a numeric matrix of size `p` by `n` containing the mixture proportions for each person in the sample.
`x`	a numeric vector of length `n` containing the observed event times for each person in the sample.
`delta`	a numeric vector of length `n` that denotes censoring (1 denotes event is observed, 0 denotes event is censored).
`timeval`	numeric value at which the distribution function is evaluated.
`qvs`	a numeric matrix of size `p` by `m` containing all possible mixture proportions (i.e., the probability of belonging to each population k, k=1,...,p.).
`p`	number of populations, must be at least 2.
`m`	number of different mixture proportions, must be at least 2.
`r`	numeric vector including the number of individuals in each mixture proportion group.
`boot`	number of bootstrap replicates.
`bootvar`	logical indicator. If TRUE, we compute the bootstrap variance estimates of the estimators.
`useOLS`	logical indicator. If TRUE, we compute the distribution function for the mixture data where the influence function is the ordinary least squares estimator.
`useWLS`	logical indicator. If TRUE, we compute the distribution function for the mixture data where the influence function is the weighted least squares estimnator.
`useEFF`	logical indicator. If TRUE, we compute the distribution function for the mixture data where the influence function is the efficient estimator.
`useNPMLEs`	logical indicator. If TRUE, we compute the distribution function for the mixture data based on the type I and type II nonparametric maximum likelihood esimators (NPMLEs).
`useEMPAVA`	logical indicator. If TRUE, we compute 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.

a list containing

hts0: numeric vector of length m containing the Kaplan-Meier estimates of the survival curves for the m mixture proportion groups.
Fest: numeric array containing the estimated distribution functions for all influence functions of interest.
var_est: numeric array containing the estimated variances for all influence functions of interest.
eflag: numeric value indicating if errors occurred in computing the estimated distribution functions. eflag<0 indicates an error.

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.

Wang, Y., Garcia, T.P., and Ma. Y. (2012). Nonparametric estimation for censored mixture data with application to the Cooperative Huntington's Observational Research Trial. Journal of the American Statistical Association, 107, 1324-1338.

tpgarcia/stride documentation built on March 18, 2021, 3:42 p.m.