reclass.mat.diag {accuracy}R Documentation

Function to produce reclassification matrices

Description

This returns a cumulative probability matrix useful for reclassifying discrete variables with PTBdiscrete.

Usage

reclass.mat.diag(n, q)

Arguments

n number of levels of factor, character vector, or discrete values
q probability of reclassifying to same sate

Details

For reclass.mat.diag, the transition probability of changing from level i to level j is q if i==j, 0 if abs(i-j)>1 and q, 1-q/2 if abs(i-j)>1 and 1>i>n and 1-q otherwise

For reclass.mat.random, the transition probability of changing from level i to level j is q if i==j, (1-q)/(n-1) otherwise.

Value

returns a matrix of cumulative probability distributions for reclassifying each level to another level

Author(s)

Micah Altman Micah_Altman@harvard.edu http://www.hmdc.harvard.edu/micah_altman/

References

Altman, M., J. Gill and M. P. McDonald. 2003. Numerical Issues in Statistical Computing for the Social Scientist. John Wiley & Sons. http://www.hmdc.harvard.edu/numerical_issues/

See Also

perturb, reclass.mat.random

Examples

  
        x<-ceiling(runif(1:100)*3)              # vector taking on 3 discrete levels
        rx <-reclass.mat.random(3,.95)  # reclassification matrix, prob of change  = .05
        rpx<-replicate(10,PTBdiscrete(x,rx),simplify=FALSE) # 100 perturbations
        if (is.R()) {
     matches <-sapply(rpx,function(y)(sum(y==x)))       # how many matches to original vector
  } else {
     # Splus variation
     matches <-sapply(rpx,substitute(function(y)(sum(y==x))))
  }
        summary(matches)                                # mean should be close to .95

[Package accuracy version 1.31 Index]