In [1]:
import pandas as pd
In [3]:
df = pd.read_csv("MBA.csv")
In [4]:
df.head(10)
Out[4]:
application_id gender international gpa major race gmat work_exp work_industry admission
0 1 Female False 3.30 Business Asian 620.0 3.0 Financial Services Admit
1 2 Male False 3.28 Humanities Black 680.0 5.0 Investment Management NaN
2 3 Female True 3.30 Business NaN 710.0 5.0 Technology Admit
3 4 Male False 3.47 STEM Black 690.0 6.0 Technology NaN
4 5 Male False 3.35 STEM Hispanic 590.0 5.0 Consulting NaN
5 6 Male False 3.18 Business White 610.0 6.0 Consulting NaN
6 7 Female False 2.93 STEM Other 590.0 3.0 Technology Admit
7 8 Male True 3.02 Business NaN 630.0 6.0 Financial Services NaN
8 9 Male False 3.24 Business White 590.0 2.0 Nonprofit/Gov NaN
9 10 Male False 3.27 Humanities Asian 690.0 3.0 Consulting NaN

Link to data set on kaggle: https://www.kaggle.com/code/mahmoudredagamail/mba-admission-dataset-class-2025/input

Question¶

Can we predict MBA acceptance at Wharton School of Business based on Gender, GPA, GMAT, work experience, and undergraduate major?

Which of these variables is most important for predicting acceptance at Wharton?

Variables¶

  • Gender: Male or Female
  • GPA: 0-4 scale
  • Work Experience: in years
  • Undergraduate Major: Business, Humanities, STEM
  • Admission: Admit or Deny

Methods¶

  1. Encoding Variables

    • Encode Gender: Male = 1, Female = 0
    • Encode Major: Business = 2, Humanities = 1, STEM = 0
    • Encode Admission: Admit = 1, Deny = 0

  2. Modeling Approaches

    • Linear Regression with Lasso: Used to identify the most important variables.
    • Decision Tree: Applied with the most significant variables remaining after Lasso selection.
In [ ]: