# What Criteria Could You Differentiate Among Multiple Discriminant Analysis

927 Words4 Pages
Questions 1. How would you differentiate among multiple discriminant analysis, regression analysis, logistic regression analysis, and analysis of variance? The main difference is in the number of independent and dependent variables and the way in which these variables are measured. Note the following definitions: In Multiple discriminant analysis (MDA), the predictor or independent variables are metric and the single criterion or dependent variable is non metric . In Regression Analysis , both the multiple independent variables and the single dependent variable are metric. In Analysis of Variance (ANOVA) , the single independent variable is non metric and the multiple dependent variables are metric. 2. What criteria could you use in…show more content…
The result is an upward bias in statistical significance that must be identified during the analysis and interpretation. 4. How would you determine the optimum cutting score? For equal group sizes, the optimum cutting score is defined by ZA + ZB ZCE = ---------- N ZCE is the critical cutting score value for equal size groups, ZA is the centroid for group A and ZB is the centroid for Group B and N is the total sample size. For unequal group sizes, the optimum cutting score is defined by NAZA + NBZB ZCU = ------------ NA + NB ZCU = is the critical cutting score value for unequal size groups, NA is the sample size for group A and NB is the sample size for Group B 5. How would you determine whether the classification accuracy of the discriminant function is sufficiently high relative to chance classification? There has to be some chance criterion that should be established. This is generally a fairly straight-forward function of the classifications that is used in the model and of the sample size. The authors then suggest the criterion that the classification accuracy or the hit ratio must be at least twenty five percent greater than by chance. Other test can be to use a test of proportions to examine for the significance between the obtained hit-ratio proportion and the chance criterion proportion . 6. How does a two-group discriminant analysis differ from a three-group