Saturday, September 15, 2012

Day 9_TEAM_D_Descriptive Analysis



What is Discriminant Analysis?

- Discriminant Analysis may be used for two objectives: either   we want to assess the adequacy of classification, given the group memberships of the objects under study; or we wish to assign objects to one of a number of (known) groups of objects. Discriminant Analysis may thus have a descriptive or a predictive objective.
The aim of all descriptive techniques is to generate quantitative data which describes the similarities and differences among a set of products. Each method has a different approach; however the basic framework of all the techniques is the same:
  • selection of panel members
  • term generation (or selection of appropriate lexicon)
  • concept formation
  • testing of panel consonance
  • evaluation of products
Descriptive techniques include Free Choice Profiling (FCP), the Spectrum™ method, Quantitative Descriptive Analysis™ (QDA), Flavor Profile Method, Texture Profile Method, Flash Profiling and generic descriptive analysis. Generic descriptive analysis generally takes pieces from QDA and Spectrum™ methods, but is modified to suit the goals of the project and limitations of the product being tested. Of the methods mentioned here, FCP and Flash Profiling involved the use of untrained consumers rather than a trained panel (although a trained panel can be used). This main point of differentiation makes these techniques faster and cheaper to conduct as there is no training involved

DA is used when:
The dependent is categorical with the predictor IV’s at interval level such as age, income,  attitudes, perceptions, and years of education, although dummy variables can be used as predictors as in multiple regression. Logistic regression IV’s can be of any level of measurement. 

There are several purposes of DA:
  • To investigate differences between groups on the basis of the attributes of the cases,
  • Indicating which attributes contribute most to group separation. The descriptive technique successively identifies the linear combination of attributes known as canonical
  • Discriminant  functions (equations) which contribute maximally to group separation.
  • Predictive DA addresses the question of how to assign new cases to groups.
  •  The DA function uses a person’s scores on the predictor variables to predict the category to which the individual belongs.
  • To determine the most parsimonious way to distinguish between groups.
  • To classify cases into groups. Statistical significance tests using chi square enable you
  • To see how well the function separates the groups.
  •  To test theory whether cases are classified as predicted
 
 
Abishek Machama
Team -D

No comments:

Post a Comment