[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: HosmerLemeshow C and H using matrix
I am in the process of comparing a published logistic regression model
to my own data. I would like to calculate H-L statistics C and H using
stata given only the beta coefficients of the published model and my
I have been able to calculate the C statistic using the following
command with Stata8.2:
lfit death, group(10) table beta(mat)
where mat is the matrix of published coefficients (col names =
variable names), and death is the dependent dichotomous variable.
Does anyone know how to get stat to calculate the H statistic using
only published coeff?
* For searches and help try: