Scribd
Upload
8 views3 pages

Conjoint Analysis and Part-Worth Utility

The document discusses utility in conjoint analysis and various functions related to calculating utilities from preference data. It outlines functions to calculate part-worth utilities, total utilities, attribute importance, utility of attribute levels, participation probabilities using different models, and clustering respondents. The functions take preference and profile data as arguments and return various utility and importance metrics.

Uploaded by

abhishek.baidya
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
8 views3 pages

Conjoint Analysis and Part-Worth Utility

The document discusses utility in conjoint analysis and various functions related to calculating utilities from preference data. It outlines functions to calculate part-worth utilities, total utilities, attribute importance, utility of attribute levels, participation probabilities using different models, and clustering respondents. The functions take preference and profile data as arguments and return various utility and importance metrics.

Uploaded by

abhishek.baidya
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Utility encompasses all features of object both tangible and

intangible as such it is an overall preference of an individual :


Utility based on the value placed on each of the levels.
Utility is expressed by relationship of attributes or the
combination of attributes
Attributes<> Factors
Conjoint(y, x, z)
Arguments
y matrix of
preferences
x matrix of
profiles
z matrix of
levels names
1. Function's characteristic
2. caPartUtilities(y, x, z) { function calculates
matrix of individual levels utilities for
respondents
(with intercept on _rst place)
3. caTotalUtilities(y, x) { function calculates
matrix of theoretical total utilities for
respondents
(for n pro_les and all respondents)
4. caImportance(y, x) { function calculates
importance of all attributes. The sum of
importance
should be 100%
5. caUtilities(y, x, z) { function calculates
utilities of attribute's levels
6. Conjoint(y, x, z) { function returns part-
worth utilities of levels (model parameters
for whole
sample), vector of percentage attributes'
importance, sum of them and corresponding
charts
(barplots). The sum of importance should be
100%
7. ShowAllUtilities(y, x, z) { function returns
basic matrix of part-worth utilities (with the
in-
tercept), matrix of total utilities (for n pro_les
and all respondents) and vector of percentage
attributes' importance, with sum of them. The
sum of importance should be 100%
8. caBTL(sym, y, x) { function estimates
participation of simulation pro_les using
probabilistic model
BTL (Bradley-Terry-Luce). The sum of
participation should be 100%
9. caLogit(sym, y, x) { function estimates
participation of simulation pro_les using logit
model. The
sum of participation should be 100%
10. caMaxUtility(sym, y, x) { function
estimates participation of simulation pro_les
using model of
maximum utility (\_rst position"). The sum of
participation should be 100%
11. ShowAllSimulations(sym, y, x)
{ function returns 3 vectors of percentage
participations using max-
imum utility, BTL and logit models. The sum
of importance for every vector should be
100%
12. caModel(y, x) { function estimates
parameters of conjoint analysis model
13. caSegmentation(y, x, c=3) { function
divides respondents into three or n clusters
using k-means
method { function kmeans() from package
stats
14. Function's arguments
15. y matrix (vector) of empirical
preferences
16. x matrix of pro_les
17. z vector with level names
18. sym matrix of simulation pro_les
19. c number of clusters (optional), default
value: c=3
20. Source: own work out.
21. culates

You might also like