2013•Communication in Statistics- Theory and MethodsRequires access

Locally D-Optimal Design for a Logit Model in Discrete Choice Experiment

Habib Jafari

Abstract

Conjoint analysis is concerned with understanding how people make choice between products or services (alternatives) or a combination of product and service (choice set), so that businesses can design new products or services that better meet customers needs. In this situation, logit model (Multinomial Logit Model) has been used to calculate the probability related to choosing an alternative in a choice set with the highest utility. Then I considered several choice sets instead of one. In this article, I have used the locally D-optimal design for the combination of the level of attributes (two attributes each with two levels) to create alternatives. The optimal combination of alternatives in choice sets which help us to have a suitable choice.

About this research paper

What this paper is about

Why it matters

OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Key concepts: Multinomial logistic regression, Choice set, Discrete choice, Mixed logit, Conjoint analysis, Set (abstract data type), Computer science, Product (mathematics)