Em 24 de fevereiro de 2010 Ulrich <Ulrich.Fritsche@wiso.uni-hamburg.de> escreveu:

I use the „educational attainment and earnings“ data set from Dougherty (2007) to explain the idea of a logit (or probit) model in the classroom.

When I moved my files from EViews to gretl, I have difficulties to generate a dummy for the cases where the value for the variable s is larger than a certain threshold (let’s say 12) to estimate what Dougherty did with his “GRAD” variable in the book.

I knew how it worked in Eviews but in gretl I thought the following part of the help is relevant:

“A tip regarding dummy variables: Suppose x is coded with values 1, 2, or 3 and you want three dummy variables, d1 = 1 if x = 1, 0 otherwise, d2 = 1 if x = 2, and so on. To create these you can use the commands:

                    genr d1 = (x=1)

                    genr d2 = (x=2)

                    genr d3 = (x=3)”

 So I tried “genr dummy = (s>11)”

 Did not work out. How can I handle this?

Dear Ulrich, I think Gretl scripts / command are case sensitive, so you need to use "S". Please try this (it worked for me):

genr dummy = (S>11)

Henrique C. de Andrade
Doutorando em Economia Aplicada
Universidade Federal do Rio Grande do Sul