Dataset: Bank Marketing
Background: I chose this dataset in order to analyze the probabilistic relationship between education levels, housing, and loans.
Goals: I hope to better understand consumer behavior in terms of financial decisions and education level.
Methods: In order to achieve my goal, I utilized Spyder software to create various cross-tabulations and probabilistic calculations.
Results and Analysis:


I chose to calculate two joint probabilities and two conditional probabilities: P(Primary Education, Housing), P(Tertiary Education, Loan), P(Housing | Primary Education), P(Loan | Tertiary Education). The marginal probabilities calculated are as follows: P(Housed), P(Primary Education), P(Loan), and P(Tertiary Education).
P(Housed)= 0.4731, meaning 47.31% of individuals in this dataset have access to housing.
P(Primary Education)= 0.1344, meaning 13.44% of individuals in this dataset have a primary education level
P(Loan)= 0.1308, meaning 13.08% of people in this dataset have taken out a loan
P(Tertiary Education)= 0.3305, meaning 33.05% of people in this dataset have a tertiary education level
P(Primary Education, Housing)= 0.0665, meaning 6.65% of people in this dataset have a primary education level and have access to housing
P(Housing | Primary Education)= 0.0495, this means that in this dataset, 4.95% of people have housing given that they a primary education level.
P(Loan, Tertiary Education)= 0.0325, meaning that 3.25% of people in this dataset have a tertiary education and have taken out a loan
P(Loan | Tertiary Education)= 0.0984, meaning that the probability that someone has taken out a loan given that they have a tertiary education level is 9.84%
Future Directions: I would be interested in further exploring the relationships between housing and marital statuses.
Leave a Reply