components: Function to get the likelihood contribution of different...

In matthiasspeidel/hmi: Hierarchical Multiple Imputation

Description

It is needed in the imputation routine for rounded income. See equation (5) in Drechsler, Kiesel & Speidel (2015)

Usage

components(ki, kj, mean1_obs, mean2_obs, sigma1, sigma2, rho, inc_obs,
  mean.log.inc, sd.log.inc, half.divisor)

Arguments

ki	An integer for the i-th treshold (or "cutpoint")
kj	An integer for the j-th treshold (or "cutpoint") (ki < kj)
mean1_obs	A vector with the expected value of G for the observed data
mean2_obs	A vector with the expected value of log(Y) for the observed data
sigma1	A scalar for the variance of the G
sigma2	A scalar for the variance of log(Y)
rho	A scalar from [-1, 1] specifying the correlation between G and log(Y)
inc_obs	The vector of the target variable (with all NAs removed)
mean.log.inc	A scalar specifying the mean of the logarithm of the target variable
sd.log.inc	A scalar specifying the standard deviation of the logarithm of the target variable
half.divisor	A scalar needed to find the bounds of possible rounding. E.g. if rounding happens to the closest multiple of 500, the half.divisor is 250.

Value

A vector with the contribution to the likelihood of every individual with an observed target variable value

References

Joerg Drechsler, Hans Kiesl, Matthias Speidel (2015): "MI Double Feature: Multiple Imputation to Address Nonresponse and Rounding Errors in Income Questions", Austrian Journal of Statistics, Vol. 44, No. 2, http://dx.doi.org/10.17713/ajs.v44i2.77

matthiasspeidel/hmi documentation built on Aug. 18, 2020, 4:37 p.m.