Building a Scorecard using Python
Financial Scorecards are used widely in all financial organizations for different kinds of ratings. This workshop will take you through the building and validation process of a financial scorecard using data.Financial Scorecards are used by banking organizations to judge the financial stability of their portfolio and take business decisions. These scorecards help in tracking and collections.
This workshop is designed for audience to take them through the process of developing a scorecard using Python. The workshop will guide you through the EDA process using Python . We would cover basics of EDA and how python visualizations can support us in data mining. We aim to cover step by step process of building a scorecard and Use of different Machine Learning algorithms to build a better scorecard by comparing the outputs of different algorithms. We will demonstrate 3 different Machine learning algorithms Random Forest , Support Vector Machine and Gradient Boosting and their outcomes while building this scorecard.
Along the workshop we would introduce you to Python libraries that can be used to build these scorecards with more efficacy.
The key python libraries that we will be using will be Pandas , Numpy ,Scipy , Matplotlib and seaborn. We would demonstrate functions of these libraries used in building scorecards.
This will be a hands on session and attendees can come with their laptops for better understanding and follow up of session.
Outline/Structure of the Workshop
Introduction to application of Financial Scorecard. What business problem does it solve ( 8-10 mins)
Exploratory Data Analysis - Data Wrangling using Python ( 18- 20 mins)
Intro to ML and Supervised Learning ( 4- 5 mins)
Understanding Random Forest ( 4-5 mins)
Building the scorecard using Machine Learning algorithms - Random Forest Walk through in Python (15 -20 mins)
Exercise with class on Implementing Gradient Boosting ( 10 mins)
Comparing Scorecards and selecting the best ( 5 mins)
Scorecard Validation ( 5 mins)
Q & A ( 10 mins)
Data wrangling and mining using Python
Building Supervised Machine Learning algorithms with algorithm understanding - Random Forest and Gradient Boosting
Compare the outputs of different algorithms.
Work with key Python libraries used for Data Analysis & Machine Learning
Understand the mathematics behind creating a perfect Financial Scorecard
Understanding of concepts like Binning , Variable Selection , IV (Information Value) , Weight of Evidence(WoE )
Machine Learning Practitioners , Aspirational Data Scientists , Data Science Enthusiast, Python Programmers , Financial Experts , Banking Professionals
Prerequisites for Attendees
Basic Knowledge of Data Science and Python.
For tools, we will be using Google Colab Jupyter notebooks with source code hosted on Github.
Alternatively, Anaconda python Distribution (Jupyter Notebooks) can also be used on local machine based on preference.
Anaconda Installation Instructions. Please install ahead of the workshop since this will not be covered during the workshop. Visit https://www.anaconda.com/distribution/#download-section ; Download the latest 3.x version.
Here is the github link for Data and Python file.
Github and google account credentials would be needed to access and save files during the workshop