gps.method.estimation: Generalized Propensity Score Estimation
In fadedsoul/cdml: CDML

Generalized Propensity Score Estimation

1
2
3

gps.method.estimation(data, gps.method = "series", n.prop = 0.7,
  bw.length.out = 10, bw.from = 0.5, bw.to = 9.5, treatment.min = -10,
  treatment.max = 10, verbose = TRUE, detoured = FALSE)

`data`	data
`gps.method`	methods for generalized propensity score estimation, choose from "series", "linear&normal", "boxcox"
`n.prop`	default = 0.7 meaning 70% for training usage, 30% for validation usage in method "series"
`bw.length.out`	number of tested points of tuning parameters from the kernel (bandwidth) in method "series"
`bw.from`	lower bound of tuning parameters from the kernel (bandwidth) in method "series"
`bw.to`	upper bound of tuning parameters from the kernel (bandwidth) in method "series"
`treatment.min`	in what interval conditional density shall be evaluated
`treatment.max`	in what interval conditional density shall be evaluated
`verbose`	print in console the procedure step by step
`detoured`	if FALSE method will be done normally; otherwise a special simulation will be examined, where CDML is chosen method, CTE is chosen model, gps.method "series" is not allowed.