Python Bootcamp for Actuaries

February 27 - March 1, 2019

AGENDA (Subject to change)

 Download PDF version of Agenda here.

Wednesday, February 27, 2019 

9:00am

Registration & Breakfast

9:30am

Opening Remarks

10:00am


Help Desk 

Come early and get help installing the tools needed (Anaconda, PyCharm, …) to be up and running right when Bootcamp starts.


Machine Learning Help Desk 

Get help installing the Tensorflow-Keras stack and running examples to make the most out of the ML sessions.

12:00pm

 Lunch

 

1:00pm

Plenary Session #1:


Actuary of the Future with Erik Wenzel, FSA, CNA Insurance 

2:00pm

 

 

Breakout Session #1:

 

Beyond Excel for Actuarial Models

Practical examples of how to take the first steps toward adding Python to your actuarial toolkit.


Creating a Web Application

For functionality that is frequently re-used and needs to be accessed by many stakeholders web applications often make more sense than shared spreadsheets.  From basic web sites in flask to fully functional platforms written in Django, Python makes it straightforward to take a single user process and turn it into a repeatable, multi-user tool.


3:00pm

 

 

Breakout Session #2:

 

APIs and Datastores 

One of the most powerful uses of Python is as a framework for integrating disparate data sources. In this session we will gather data from publicly available APIs and then combine and manipulate it with data from a local datastore.  

Making Python Code Really Fast

Practical examples of optimizing Python code for best performance; introduces Cython.

4:00pm

 

 

Breakout Session #3:

 

Web Scraping 

Sometimes data isn’t in a convenient location. In this session we give useful examples of scraping data from websites and other documents. 

Lessons in Library Building

Actionable best practices in building reliable, well-tested, reusable, and sharable packages in Python.

5:00pm

Closing Remarks 

 

Thursday, February 28, 2019

9:00am

Breakfast

9:55am

Opening Remarks 

10:00am

Plenary Session #2:


Explainable Deep Learning with Slater Victoroff 

11:00am

 Breakout Sessions #4:

Introduction to Machine Learning in Python

A practical introduction to using industry standard machine learning libraries in Python and Jupyter.  Includes example code.

 

Monte Carlo in Python

We show how to simulate basic models (e.g. Black-Scholes and CIR) as well as multi-dimensional copula models.  We then demonstrate how to manipulate and transform large matrices of simulated data.


12:00pm

 Lunch

1:00pm 

Plenary Sessions #3:

Building Powerful Tools with XL Wings

Excel is a core platform for both financial and insurance applications. XLWings makes it possible to integrate the big data capabilities of Python with the ease and familiarity of Excel reporting and visualization.


2:30pm

Breakout Sessions #5:

Visualization in Python

From the full flexibility of Matplotlib’s PyPlot to the sleek modern look of the Seaborn package, Python makes it possible to produce attractive, clear visualizations without sacrificing information density.


Estimating VaR with Importance Sampling

VaR – and analogous risk measures – is a key indicator in portfolio and investment management.  This session will teach you how to implement the importance sampling framework in Python in order to measure risk.


3:30pm

Coffee Break

4:00pm 

Breakout Sessions #6:

Pyliferisk

While the use of Python in the actuarial community is not pervasive, there are a few very useful, actuary specific libraries.  In this session we demonstrate uses of Pyliferisk.

 

Working with Text Embedding: Introduction

Basic hands-on introduction to embeddings.


4:30pm

Panel Discussion

AI, ML, and Insurance with Michael Scullard, Greg Meena, Hongjun Ha, and Slater Victoroff

5:00pm

Closing Remarks and Highlights

6:00pm

Dinner at Breckenridge Brewery

 

Friday, March 1, 2019

9:00am

Breakfast

9:30am

Opening Remarks

10:00am

Plenary Session #4:


Advice for a Long and Happy Actuarial Career with Bill Gorski 

11:00am

Breakout Sessions #7:

Estimating a Probability of Large Loss with the Least-Squares Monte Carlo and Importance Sampling

A step-by-step Python example of calculating capital requirements based on Least Squares Monte Carlo.


Machine Learning 2

Advanced machine learning applications using large, commonly available datasets.  


12:00pm

Closing Luncheon

 Download PDF version of Agenda here.

 

Register Here →