View source: R/gheckman.R View source: R/gheckman — копия.R
gheckman | R Documentation |
Use this function to estimate Multivariate Switch model
gheckman(data, outcome, selection1 = NULL, selection2 = NULL, selection3 = NULL, selection4 = NULL, selection5 = NULL, group = NULL, zo3 = NULL, ShowInfo = TRUE, onlyTwostep = FALSE, opts = list(algorithm = "NLOPT_LD_TNEWTON", xtol_rel = 1e-08, print_level = 1, maxeval = 1e+06), x0 = NULL)
data |
data frame containing the variables in the model |
outcome |
main (continious) equation formula |
selection1 |
the first selecton equation formula |
selection2 |
the second selecton equation formula |
selection3 |
the third selecton equation formula |
selection4 |
the fourth selecton equation formula |
selection5 |
the fifth selecton equation formula |
group |
vector which determines outcome to each group |
zo3 |
matrix which rows correspond to possible combination of selection equations values. |
ShowInfo |
shows likelihood function optimization info |
onlyTwostep |
if true then only two-step procedure used |
opts |
options that are passed to nlopt |
x0 |
optional initial values to be used while solving optimization task |
This function estimates Multivariate Switch model via maximum-likelihood and two-step procedures. This model was developed by Kossova E.V. and Potanin B.S. Dependent variables in selection equations should have values -1 or 1. Also there is a special value 0 which indicates that this observation is unobservable but it is necessary to take into consideration other selection equation information while calculating likelihood function. The i-th row of zo3 corresponds to i-th element of group. zo3 rows should contain information regarding possible combinations of selection equations values. While group determines the outcome for each of this possible combinations. Special 0 value for group responsible for sample selection.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.