Hi everyone, this is a first time I've try to use DCE for my paper by using dcreate command in Stata. But I have some questions:
1. This is my code in R(DCEtool package) and Stata(dcreate command):
Then, I got D-efficiency (from Stata): D-efficiency after iteration 3: 2.0285077441 & D-error (from R): 0.9355574. Because both of them create optimal design (D-optimal) so I confuse why these two are difference so which one is correct?
2. How can we identify the Dominant Alternatives generated by the software using code? If, after manually removing the Dominant Alternatives, do I need to re-verify the D-eff value, and if so, how?
1. This is my code in R(DCEtool package) and Stata(dcreate command):
In R:
Number of attributes: 5
Levels per attribute: 3 3 3 2 2
Alternatives per choice set: 2
Number of sets: 16
Null alternative: TRUE
Bayesian priors: FALSE
Priors: 0 0 0 0 0 0 0 0 0
Attribute names: price co2 marine fish food
Level names: 1 2 3 1 2 3 1 2 3 1 2 1 2
Details: Fedorov modified algorithm used to generate an optimal design (idefix >= 1.1.0)
Number of attributes: 5
Levels per attribute: 3 3 3 2 2
Alternatives per choice set: 2
Number of sets: 16
Null alternative: TRUE
Bayesian priors: FALSE
Priors: 0 0 0 0 0 0 0 0 0
Attribute names: price co2 marine fish food
Level names: 1 2 3 1 2 3 1 2 3 1 2 1 2
Details: Fedorov modified algorithm used to generate an optimal design (idefix >= 1.1.0)
In Stata:
matrix levmat = 3,3,3,2,2
genfact, level(levmat)
matrix output=J(1,5,1)
matrix b=J(1,9,0)
dcreate i.x1 i.x2 i.x3 i.x4 i.x5, nalt(2) nset(16) fixedalt(output) asc(3) bmat(b)
rename x1 price
rename x2 co2
rename x3 marine
rename x4 fish
rename x5 food
matrix levmat = 3,3,3,2,2
genfact, level(levmat)
matrix output=J(1,5,1)
matrix b=J(1,9,0)
dcreate i.x1 i.x2 i.x3 i.x4 i.x5, nalt(2) nset(16) fixedalt(output) asc(3) bmat(b)
rename x1 price
rename x2 co2
rename x3 marine
rename x4 fish
rename x5 food
2. How can we identify the Dominant Alternatives generated by the software using code? If, after manually removing the Dominant Alternatives, do I need to re-verify the D-eff value, and if so, how?
