PROJECTS :
Bank Loan Performance Analyst
📊 Bank Loan Performance Analyst | Data-driven Insights | Strategic Decision-Making
As a seasoned data analyst, I spearheaded a comprehensive Bank Loan Performance Analytics Project, leveraging Excel, SQL, and Power BI. Here’s what I achieved:
Strategic KPI Development:Created robust metrics for total loan applications (38.6k), funded amounts ($435.8M), received amounts ($473.1M), average interest rates (12.0%), and debt-to-income ratios (13.3%).
Conducted Month-to-Date (MTD) and Month-over-Month (MoM) analyses, providing actionable insights for executive decision-makers.
Good vs. Bad Loan Analysis:Defined and meticulously tracked KPIs for good loans (86.2%) and bad loans (13.82%).
Enhanced risk management strategies and empowered data-driven lending decisions.
Comprehensive Loan Status Grid:Engineered a dynamic grid to categorize loan statuses, enabling real-time monitoring and informed actions.
Unlocked critical insights into loan performance and portfolio health.
Advanced Trend and Regional Analysis:Developed intuitive visualizations to identify lending trends across regions
Comprehensive Banking Insights Initiative (CBII)
In this SQL project, I developed and executed a series of complex SQL queries on a banking database to extract actionable insights and support strategic business decisions. Key achievements include:
-
Customer Segmentation: Identified all customers from Italy who joined since 2015, aiding targeted marketing strategies.
-
Loan Analysis: Calculated the total number of loans per country to understand geographical credit distribution and identified customers with multiple loans to assess credit risk and customer loyalty.
-
Banker Performance: Evaluated the workload and customer interaction of bankers, distinguishing those who have never been assigned a loan to identify gaps in resource utilization.
-
Financial Insights: Derived the total loan amount per customer and average property values by country, crucial for assessing market value trends and individual credit exposure.
​
data visualization ( python )
This Python project involves a comprehensive analysis of two-year sales data from a pharmaceutical company, focusing on various regions, customer segments, and time frames. The project leverages Python’s data manipulation library, Pandas, along with visualization libraries such as Matplotlib or Seaborn, to uncover trends and insights from 2015 and 2016 sales data.
Project Summary:
-
Data Transformation: The data set is transformed using techniques like pivoting (wide to long and vice versa) and aggregation with groupby() to facilitate detailed analysis.
-
Sales Comparison by Region: Utilizes a bar chart to compare the sales for each region across 2015 and 2016, highlighting growth or decline in different geographical areas.
-
Factor Analysis for 2016 Sales: A pie chart representation helps identify contributing factors to 2016 sales for each region, revealing the impact of different variables on sales performance.
-
Sales Analysis by Region and Customer Tier: Compares total sales for the years 2015 and 2016, segmented by region and customer tiers, providing insight into customer behavior across different segments.
-
State-Level Performance in the East Region: Identifies states within the East region that registered a decline in sales from 2015 to 2016, pinpointing areas of concern where sales strategies might need adjustment.
-
Performance in High Tier Divisions: Analyzes divisions within the High customer tier to determine where there was a decrease in unit sales in 2016 compared to 2015.
-
Quarterly Sales Analysis: A new column for quarters is created using numpy.where() or similar functions based on months, facilitating a quarterly analysis of sales. Sales data for both years is then compared using a bar plot for each quarter.
-
Tier Composition in Quarterly Sales: Four pie charts are created to visualize the composition of quarterly sales in 2016, dissected by different customer tiers, providing a detailed breakdown of sales contributions by customer segment.
​
