narrow the results to a legible Dendrogram. I can then look at the clustered variables to determine a correlation. K-Means Clustering The K-Means Cluster analysis allows you to define several “centroid” clusters and all of the data in the sample is distributed to those clusters by “means” or averaging the information.
Linear Regression The simple linear regression analysis allows us to see if two variables have a relationship with one another by testing one variable against the other and showing a probability of a correlation.
Logistic Regression The logistic regression analysis will compare the probability of a correlation between one dependent binary variable and one or multiple independent variables. In analyzing these variables there are one of two outcomes Yes or no, 1 or 0, true or false, etc. In the examples above I test web purchases against income, age, marriage, res_visits, and marriage by third visits to determine a correlation. We can see that married couples frequent the restaurant more often and tend to make less web purchases. Younger people tend to make more web purchases, and lower and average income tend to make more web purchases. It is limited to those binary results and specific types of variables.
You've reached the end of your free preview.
Want to read all 6 pages?