Consumer Loan Portfolio Analysis and Risk Assessment
Stack: #PostgreeSQL #sqllite , Microsoft #Excel #PowerQuery
What was done:
Data Processing ( #etl ): Cleaned and structured a "raw" dataset (German Credit Data) of 1000 records.
#PostgreeSQL: Wrote a series of queries using CTE, and Case When to segment customers by risk level, age, and loan purpose.
Excel Modeling: Built an automated table using XLOOKUP and IFS functions to decipher categorical codes. After that, pivot tables were created to calculate the average check and the percentage of defaulted loans by housing conditions and professions.
An analytical report has been generated that identifies the highest-risk segments of borrowers (for example, young people with education loans), which allows for optimizing credit policy.
What was done:
Data Processing ( #etl ): Cleaned and structured a "raw" dataset (German Credit Data) of 1000 records.
#PostgreeSQL: Wrote a series of queries using CTE, and Case When to segment customers by risk level, age, and loan purpose.
Excel Modeling: Built an automated table using XLOOKUP and IFS functions to decipher categorical codes. After that, pivot tables were created to calculate the average check and the percentage of defaulted loans by housing conditions and professions.
An analytical report has been generated that identifies the highest-risk segments of borrowers (for example, young people with education loans), which allows for optimizing credit policy.
Kropivnitskiy 
Available for hire