Conjoint analysis

From DDL Wiki

Conjoint analysis, also known as multiattribute compositional models.

Contents 1 Basic Process 2 Data Collection Methods 3 Data Analysis 4 Example 5 External links 6 Reference

Basic Process

The basic process for a conjoint analysis is as follows:

Data Collection Methods

• Rating (e.g. rating with a scale from 1-10)
• Ranking (e.g. rank best as 1st, 2nd, 3rd, etc.)
• Choice (e.g. four options, which do you choose?)

The conjoint analysis can be done with different types of data, which include stated choice and revealed preference. The differences between the two are outlined below:

• Stated choice - Using data from a survey

  Pros: - controlled experiment
  Cons: - survey item choices are not always the same as what is desired or already in the marketplace

• Revealed Preference - Using data collected from actual results in the marketplace

  Pros: 

Surveys can be generated by using SAS to get full-factorial and partial-factorial results.

Data Analysis

logit and probit

Example

External links

• DONLP2 is an open-sourced code of implementing SQP algorithm. The website offers packages in Fortran 77, F90 and ANSI-C versions.

Reference

• G.N. Vanderplaats, Numerical Optimization techniques for Engineering Design with Applications, 1984, McGraw-Hill Inc.

• P.Y. Papalambros and D.J. Wilde, Principles of Optimal Design: Modeling and Computation, 1988, Cambridge University Press.

• S.S. Rao, Engineering Optimization: Theory and Practice, 1996, 3rd Ed., John Wiley & Sons, Inc.