[Limdep Nlogit List] Welfare Calculation using Simulation Commands

[David W. Carter]

I am trying to calculate the mean welfare effect of changing an 
attribute in a conditional logit model using the inclusive values 
produced in simulation commands. Would the following code sketch 
estimate the average welfare effect of changing attribute1 from 4 to 2, 
assuming the parameter on attribute2 measures the marginal utility of 
income? Thanks.

NLOGIT   ; Lhs= y,nij,altij
                  ; Choices = alt1,alt2,alt3
                  ; Rhs = attribute1, attribute2$

NLOGIT   ; Lhs= y,nij,altij
                  ; Choices = alt1,alt2,alt3
                  ; Rhs = attribute1, attribute2
                  ; Simulation
                  ; Scenario: attribute1=4
                  ; Ivb=ibv4$

NLOGIT   ; Lhs= y,nij,altij
                  ; Choices = alt1,alt2,alt3
                  ; Rhs = attribute1, attribute2
                  ; Simulation
                  ; Scenario: attribute1=2
                  ; Ivb=ibv2$

CREATE   ;  cv42 = (ivb4-ivb2)/B(2)$

REJECT    ; y = 0$
DSTAT     ; Rhs = cv42$


-- 
David W. Carter, Ph.D.
Southeast Fisheries Science Center
75 Virginia Beach Drive
Miami, Florida 33149
Tel. (305) 361-4467
Fax. (305) 365-4102
david.w.carter at noaa.gov

"The contents of this message do not necessarily reflect
any position of the Government or the National Oceanic and Atmospheric Administration."