Examine how the market structure impacts the company

Recall that the single dependent variables in regression are the appropriate statistical techniques when the research problem involves a single categorical dependent variable and several metric independent variables. In many cases, the dependent variable consist of two groups or classifications, for example, male versus female, high verses low, or good versus bad. In other Instances, more than two groups are Involved, such as low, medium, and high classifications. Discriminate analysis and logistic regression are capable of handling either two groups or multiple (three or more) groups.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

The results of a discriminate analysis and logistic regression can assist In profiling the interrupt characteristics of the subjects and in assigning them to their appropriate groups. 2. Discuss the major issues relating to types of variables used and sample size required in the application of discriminate analysis. Chapter 5 100 word minimum. To apply discriminate analysis, the researcher first must specify which variables are not to be independent measures and which variable is to be dependent measure.

The researcher should focus on the dependent variable first. The number of dependent variable groups must be mutually exclusive and exhaustive. After a decision has been made on the dependent variable, the researcher must decide which Independent variable to include In the analysis. Independent variables are selected In two ways: 1 . By Identifying variables either from previous research or from the theoretical model underlying the research question, and 2.

By utilizing the researcher’s knowledge and Intuition to select variables for which no previous research or theory exists but that logically might be related to predicting the dependent variable groups. Discriminate analysis like the other multivariate techniques is affected by the size of the sample being analyzed. A ratio of 20 observations for each predictor variable is recommended. Because the results become unstable as the sample size decreases relative to the number of independent variables, the minimum size recommended is five observations per independent variable.

The sample size of each group also must be considered. At a minimum, the smallest group size of a category must exceed the number of independent variables. As a practical guideline, each category should have at least 20 observations. Even If all categories exceed 20 observations, however, the researcher also must consider the relative size of the groups. Wide variations In the size of the groups will affect the estimation of the Dalmatians function and the classification of