bamlss要点总结

核心估计函数
bamlss() 包装为 opt_bfit()

opt_bfit(x, y, family, start = NULL, weights = NULL, oiffset = NULL, ...)

默认的sampling

sam_GMCMC(x, y, family, start = NULL, weights = NULL, offset = NULL, ...)

	opt_bfit	sam_GMCMC
返回值	回归系数vector	matrix，mcmc或者mcmc.list
前缀	opt_*	sam_*

parametric part p
nonlinear smooth effect terms s

set the seed for the reproducibly

set.seed(123)

model formula

f <- ofp ~ hosp + health + numchron + gendar + school + privins

Estimate model

b1 <- bamlss(f, family = "poisson", data = DebTrivedi,
         n.iter =  12000, burnin = 2000, thin = 10)

f <- list(
ofp ~ hosp + health + numchron + gender + school + privins,
sigma ~ hosp + health + numchron  + gender + school + privins
nu ~ 1
)