Example of a full-factorial design with ChoiceDesign

This notebook illustrates how to use ChoiceDesign to generate a simple full-factorial experimental design.

Step 1: Load modules and define attributes

from choicedesign.design import FullFactDesign
from choicedesign.expressions import Attribute

Each attribute is defined with Attribute(name, levels). The following lines define 4 attributes across 2 alternatives:

alt1_A = Attribute('alt1_A', [1, 2, 3])
alt1_B = Attribute('alt1_B', [10, 15, 15.5])

alt2_A = Attribute('alt2_A', [1, 2, 3])
alt2_C = Attribute('alt2_C', [0, 3, 5])

Step 2: Construct design object and generate the design matrix

FullFactDesign takes a list of Attribute objects. gen_design() returns a DataFrame with all combinations of attribute levels (3×3×3×3 = 81 rows).

design = FullFactDesign(X=[alt1_A, alt1_B, alt2_A, alt2_C])
full_design = design.gen_design()
full_design

	alt1_A	alt1_B	alt2_A	alt2_C
0	1	10.0	1	0
1	1	10.0	1	3
2	1	10.0	1	5
3	1	10.0	2	0
4	1	10.0	2	3
...	...	...	...	...
76	3	15.5	2	3
77	3	15.5	2	5
78	3	15.5	3	0
79	3	15.5	3	3
80	3	15.5	3	5

81 rows × 4 columns