help in orthogonal or efficient design syntax

[lich93]

Dear all,
I am considering the following elements in a CE design:
3 alternatives: status quo (0), some improvement (1), and much improvement (2)
6 attributes with the same 3 levels (0,1,2), and 1 local fee attribute (Tax) with 5 levels (-10% -5%, 0%, +5%, +10%) showing percentage increase/decrease in the tax or fee for the best management option.

I would like to know whether a D-optimal (OOD) or "Orthogonal fractional factorial" design would be appropriate for my study as I do not have any Bayesian priors to conduct an efficient design.

I am facing difficulties with five levels (in percentage) for the cost attribute. However, if I use the same three levels (0,1,2) for the cost attribute to proxy for (0%, 5%, 10%), the orthogonal design seems to be okay.

I used the following designs:

Design
;alts = alt1*, alt2*, alt3
;rows = 12
;orth = seq
;eff = (mnl,d)
;block = 3
;model:
U(alt1) = b1*ATR[2,1,0] + b2*ADM[2,1,0] + b3*AFR[2,1,0] + b4*OFI[2,1,0] + b5*PLFP[2,1,0] + b6*SFM[2,1,0] + b7*TAX[2,1,0] /
U(alt2) = b1*ATR + b2*ADM + b3*AFR + b4*OFI + b5*PLFP + b6*SFM + b7*TAX
$

Design
;alts = alt1, alt2, alt3
;rows = 12
;orth = seq2
;eff = (mnl,d)
;block=3
;model:
U(alt1) = b1*ATR[2,1,0] + b2*ADM[2,1,0] + b3*AFR[2,1,0] + b4*OFI[2,1,0] + b5*PLFP[2,1,0] + b6*SFM[2,1,0] + b7*TAX[2,1,0] /
U(alt2) = b1*ATR + b2*ADM + b3*AFR + b4*OFI + b5*PLFP + b6*SFM + b7*TAX
$

I have some concerns and queries:
1. Is there any possibility to create 8 choice scenarios with the existing attributes and levels? If yes, how?
2. If not, or at least 12 scenarios are required, then can I divide the design into two blocks or three blocks so that each respondent faced with fewer choice scenarios?
3. Can I introduce a few interactions between attributes in the design? Does it improve the design performance? What are the odds?
4. How can I accommodate all five levels of the cost attribute in the design to measure marginal willingness to pay later?

I am in the initial stage of the CE design and any help will be greatly appreciated as it will be a great contribution to my Ph.D. research.

I look forward to getting a reply.

Sincerely,
Liton

[lich93]

Hi again,

I also tried with efficient designs, and the syntax is as follows:
?D-efficient Design with 5 levels in the cost attribute
Design
;alts = FVRmax, FVRsom, FVRstq
;rows = 15
;block = 3
;eff = (mnl,d)
;model:
U(FVRmax) = b1*ATR[2,1,0] + b2*ADM[2,1,0] + b3*AFR[2,1,0] + b4*OFI[2,1,0] + b5*PLFP[2,1,0] + b6*SFM[2,1,0] + b7*TAX[2,1,0,-1,-2] /
U(FVRsom) = b1*ATR + b2*ADM + b3*AFR + b4*OFI + b5*PLFP + b6*SFM + b7*TAX /
U(FVRstq) = ACSFVRstq
$

?D-efficient Design with 4 levels in the cost attribute
Design
;alts = FVRmax, FVRsom, FVRstq
;rows = 12
;block = 3
;eff = (mnl,d)
;model:
U(FVRmax) = b1*ATR[2,1,0] + b2*ADM[2,1,0] + b3*AFR[2,1,0] + b4*OFI[2,1,0] + b5*PLFP[2,1,0] + b6*SFM[2,1,0] + b7*TAX[3,2,1,0] /
U(FVRsom) = b1*ATR + b2*ADM + b3*AFR + b4*OFI + b5*PLFP + b6*SFM + b7*TAX /
U(FVRstq) = ACSFVRstq
$

?D-efficient Design with 3 levels in the cost attribute
Design
;alts = FVRmax, FVRsom, FVRstq
;rows = 12
;block = 3
;eff = (mnl,d)
;model:
U(FVRmax) = b1*ATR[2,1,0] + b2*ADM[2,1,0] + b3*AFR[2,1,0] + b4*OFI[2,1,0] + b5*PLFP[2,1,0] + b6*SFM[2,1,0] + b7*TAX[2,1,0] /
U(FVRsom) = b1*ATR + b2*ADM + b3*AFR + b4*OFI + b5*PLFP + b6*SFM + b7*TAX /
U(FVRstq) = ACSFVRstq
$

Would you please let me know if there remain any other optimal or efficient designs with less number of choice sets.
I am also confused whether I should use D-optimal / D-efficient / Orthogonal Sequential Design / Orthogonal Optimal Design (OOD).
I would like to use a CE design for data collection, which is statistically sound and efficient with less number of choice sets.
Any help in this regard will be highly appreciated.

Sincerely,
Liton