choice-metrics.com

Dear Prof. Michiel,
I run the syntax with

Code: Select all

design
;alts=alt1*,alt2*,alt3
;rows=12
;block=2
;eff = (mnl,d,median)
;con
;bdraws = sobol(5000)
;model:
U(alt1)=b0 [(n,5.51892, 2.400485)]+b1.dummy[(n,-.1969752, 0.6256593) |(n, -1.777203, 1.348568)]*mode[1,2,0]+b2.dummy[(n, -0.0287256, 0.20309630)]*format[1,0]+b3.dummy[(n,1.368347, 1.700064)]*con_ser[1,0]+b4[(n,-.0268496, .0316639)]*time[10,20,30]+ b5 [(n,-.0297015, .0246112)]*cost[0,50,100]/
U(alt2)= b0 [(n,5.51892, 2.400485)]+b1*mode+b2*format+b3*con_ser+b4*time+b5*cost
$


I select the minimun Db-error of evalution (0.196023), and I found unbalanced level only cost attribute between 2 blocks. Block 1; 0= 4 times,50 = 6 times, 100 = 2 times and Block 2; 0=4 times, 50 =2 times,100=6 times.
My question is
I should select the other evaluation that has a balanced level?


Sincerely yours,
Sukunta

Having attribute level balance is nice but not necessary. It is not possible to have perfect attribute level balance for all attributes within a block in an efficient design, this is only possible in orthogonal designs. So you will always end up with some attribute level imbalance within a block, and this is not a problem.

It is fine to make manual changes to the blocks if you think you can improve attribute level balance within each block, there is no best way of doing this. But you will often see that when you improve the balance of one attribute, another attribute becomes less balanced.

In your case, I think the blocks are fine, all levels appear at least twice in each block, so I would not worry too much about it.

MIchiel

Dear Prof.Michiel,
Thank you so much.

Sincerely yours,
Sukunta