1 0 1 1000 15
 3 3 3 3 3
 1 1 1 1 1
 2 2 2 2 2

 0 0 0 0 0
 3 3 3 3 3
 5 6 7 5 6

iestflg iacrl itype nfac nvar
nxkind(x1-x5)
nxkind(x6-x10)
nxkind(x11-x15)

nxcut(x1-x5)
nxcut(x6-x10)
nxcut(x11-x15)

Parameter Definitions:
 iestflg = 0, if subroutine simul simply can read previously estimated means,
                cut-off points and covariance matrix
         = 1, if program jamest3 must compute data moments
 iacrl
 itype   = 1, if estimating moments for private primary facilities,
         = 2, if estimating moments for public primary facilities,
         = 3, if estimating moments for private secondary facilities,
         = 4, if estimating moments for public secondary facilities.
  <YOU PROBABLY DO NOT WANT TO CHANGE THIS PARAMETER FROM "1">

 nfac    = number of observations in simulated data file
  <YOU MUST CHANGE THIS PARAMETER TO THE NUMBER OF OBSERVATIONS IN
     YOUR DATABASE>

 nvar    = number of variables included in analysis
  <YOU MUST CHANGE THIS PARAMETER TO THE NUMBER OF VARIABLES IN 
     YOUR DATABASE>

 nxkind(nvar=15):  describes the type of each variable
        = 1, if variable is binary
        = 2, if variable is ordered discrete
        = 3, if variable is continuous
  <YOU MUST ADAPT THESE INDICATORS TO THE ORDER AND TYPE OF VARIABLES
     IN YOUR DATABASE>

 nxcut(nvar=15):  equals the number of cut-off points to be estimated for
                      discrete variables
        = 0, if variable is continuous
        = 3, if variable is binary
        = number of outcomes + 1, if variable is ordered discrete
  <YOU MUST ADAPT THESE INDICATORS TO THE ORDER AND TYPE OF VARIABLES
     IN YOUR DATABASE>