W.multi.ak: general AK weights as a function of a and k parameters.

In DanielBonnery/CompositeRegressionEstimation: X

Description

general AK weights as a function of a and k parameters.

Usage

W.multi.ak(
  months,
  groups,
  S,
  S_1 = S - 1,
  ak,
  eta0 = length(groups)/length(S),
  eta1 = eta0 - 1,
  rescaled = F
)

Arguments

S	a vector of integers indicating the indices of the rotation group in the sample
ak	a list of 2-dimension vectors
nmonth	an integer, indicating number of months
ngroups	: number of groups

Value

an array of AK coefficients W[m2,m1,mis1] such that Ak estimate for month m2 is sum(W[y2,,])*Y) where Y[m1,mis1] is direct estimate on mis mis1 for emp stat y1 at month m1.

Examples

W.multi.ak(months=1:3,groups=1:8,S=c(2:4,6:8),ak=list(c(a=.2,k=.5),c(a=.2,k=.4))) 

DanielBonnery/CompositeRegressionEstimation documentation built on June 17, 2020, 12:16 p.m.