Explore BrainMass

Explore BrainMass

    Creating suitable regression model

    This content was COPIED from BrainMass.com - View the original, and get the already-completed solution here!

    Using any data of your choice or use the attached: build an optimized regression model eliminating non-significant Betas. Provide the Betas for the significant variables.

    How can a institution use these results to optimize its cost and labor in acquiring new students (enrollments)?  How can one use these results to sharpen the institution brand image against other institutions or known competitors?

    © BrainMass Inc. brainmass.com June 4, 2020, 5:30 am ad1c9bdddf


    Solution Preview

    Please see the attachments for the full Solution.

    1. Among the list of variables, we take the following steps to clean the dataset, making them suitable for regression analysis.
    - Create a binary dependent variable "Enrollment," which equals 1 when "Enroll" is "TRUE;" and 0 otherwise.
    - Create a "Women" binary variable for those whose "Gender" is "F"
    - Create a "Non-White" binary variable for those whose "Ethnicity" is NOT "White"
    - Create an "Age 19 or above" binary variable for those whose age is 19 or above

    - Create a "Suburb" binary variable for those whose "Region description" is in an Suburban area
    - Create a "Other area" binary variable for those whose "Region description" is outside of Chicago
    - This makes "Chicago" the default group in the regression analysis

    - Create three binary variables for "Act category description":
    "24-27"; "28-32", and "Unknown."
    Therefore, the "16-23" category becomes the default group

    - Create six binary variables for "Colleges":
    "Communication", "Computer Science", "Liberal Arts & Social Sciences",
    "Science and Health", "Music", "Theatre Arts"
    Therefore, "Commerce" becomes the default group.
    - Create "No_first_aid" for those whose "Fin_aid_interest" is ...

    Solution Summary

    The solution provides detailed explanation about regression analysis. It shows the steps to building regression models: cleaning the raw data; creating new variables; selecting the model; making decision based on statistical significance; and interpreting the economic meaning of regression results.