nlme | R Documentation |
function to compute nonlinear models with two sided measurement error
nlme(
data,
Yformla,
Tformla,
tau,
tvals,
copType = "gaussian",
simstep = "MH",
ndraws = 250,
reportTmat = TRUE,
reportSP = TRUE,
reportUM = TRUE,
reportPov = TRUE,
povline = log(20000),
reportQ = c(0.1, 0.5, 0.9),
Ynmix = 1,
Tnmix = 1,
tol = 1,
iters = 400,
burnin = 200,
drawsd = 4,
messages = FALSE,
se = FALSE,
biters = 100,
cl = 1
)
data |
data.frame |
Yformla |
formula for outcome model |
Tformla |
formula for treatment model |
tau |
values of tau to estimate first step quantile regressions for |
tvals |
values of the treatment to compute conditional distribution-type parameters for |
copType |
what type of copula to use in second step. Options are "gaussian" (the default), "clayton", or "gumbel" |
simstep |
whether to use an MH algorithm ("MH") or an importance sampling algorithm ("ImpSamp") |
ndraws |
number of draws to use in MH algorithm to estimate first step quantile regressions (default 250) |
reportTmat |
whether or not to report a transition matrix |
reportSP |
whether or not to report Spearman's rho (rank-rank correlation) |
reportUM |
whether or not to report upward mobility parameters |
reportPov |
whether or not to report fraction of population below the poverty line as a function of parents' income |
povline |
value of the poverty line (default log(20000)) |
reportQ |
quantiles of child's income as a function of parents' income to report (default is .1,.5,.9) |
Ynmix |
number of mixture components for outcome measurement error model |
Tnmix |
number of mixture components for treatment measurement error model |
tol |
tolerance for first step quantile regression model to converge
(default is 1). Note that convergence will be sensitive to |
iters |
the number of MCMC iterations (default is 400) |
burnin |
the number of MCMC iterations to drop (default is 200) |
drawsd |
starting value of standard deviations of mixture components |
messages |
whether or not to report details of computation as they
occur (default is |
se |
whether or not to estimate standard errors using the boostrap (default is FALSE) |
biters |
if computing standard errors, the number of bootstrap iterations to use (default is 100) |
cl |
allows for parallel processing in computing standard errors using the bootstrap (the default is 1) |
list of nonlinear measures of intergenerational income mobility adjusted for measurement error
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.