10 Reasons Why Every Actuary Should Know Python – MuSigma 2021-04

Speakers: Jeff White and Alex Walrath, Gross Consulting

Description:

In 2021, Excel remains the tool of choice for actuaries.  Most actuarial processes and analyses are managed in our many, complex spreadsheets.  We often stretch the limits of this tool, and why not, you might ask?  Excel is very versatile and user friendly.

In this age of digital everything and copious amounts of data, we believe the time is right to expand your toolkit.  But to what?  There are many competing and compelling choices.  But as the title suggests, we think the time is right to commit to Python.  This webinar will cover 10 reasons why Python should be your second tool and include practical demonstrations of Python in use.

Popularity indexes

https://www.tiobe.com/tiobe-index/

https://pypl.github.io/PYPL.html

https://redmonk.com/sogrady/2021/03/01/language-rankings-1-21/

https://octoverse.github.com/

Training / Resources

https://docs.python-guide.org/

https://www.w3schools.com/python/default.asp

https://realpython.com/

https://www.askpython.com/

https://www.freecodecamp.org/learn/data-analysis-with-python/

Books

https://fbeedle.com/our-books/23-python-programming-an-introduction-to-computer-science-3rd-ed-9781590282755.html

https://realpython.com/products/python-basics-paperback/

Examples

jeffdwhite/10-Reasons-Python (github.com)

 

 

Individual Claim Development Models and Detailed Actuarial Reserves in Property-Casualty Insurance

Individual Claim Development Models and Detailed Actuarial Reserves in Property-Casualty Insurance

Abstract

Actuarial reserving techniques using aggregated triangle data are ubiquitous in the property casualty insurance industry. By instead starting with the modeling of individual claim behavior using predictive modeling techniques and a modeling framework that describes the full life cycle of a claim, there are numerous benefits including greater reliability of reserve estimates, faster recognition of underlying mix changes, and avoidance of problems in pricing due to differences in development. Component development and emergence models used in conjunction with simulation of currently outstanding claims and simulation of claims still yet to be reported form an alternative framework for generating estimates of reserve need. Algorithmic case reserves at the claim level and algorithmic IBNR estimates at the policy level, actuarially determined and designed to be unbiased, provide valuable information for downstream analyses, a bridge to the generally accepted triangle reserving paradigm, and a means for demonstration of reliability for actuarial purposes.

Author: Chris Gross – Chief Executive Officer at Gross Consulting

 

Towards A Holsitic Approach for Managing Reserve Risk – MuSigma 2021-03

Speaker: Kevin Madigan, Gross Consulting

Description:

“Reserve risk” is a complicated concept, meaning different things to different stakeholders. Corporate actuarial and finance departments focus on the variability around the estimated mean of the liabilities and the likelihood that claims emergence will be significantly different from the selected accounting entries – and its impact on surplus. Pricing actuaries and underwriters focus on the risk that the expected underwriting experience may be misestimated due to misestimation and misunderstanding of similar recent underwriting experience. Capital modelers focus on the tail of the liability distributions to identify capital charges and assess capital adequacy. Claims departments focus on legal or operational risks associated with claims handling procedures.

So, some clarity is called for when actuaries use the phrase “reserve risk”. One interesting observation that arises from the above distinction is that “managing reserve risk” is also in the eye of the beholder. In this webinar Kevin will discuss how a clear understanding of an insurer’s claims, underwriting, and actuarial capabilities coupled with a clearly articulated and practical risk appetite framework can help insurers reduce the likelihood of unexpected reserve development, and help them arrive at an underwriting strategy that is more likely to produce a portfolio whose risk characteristics are aligned with corporate goals.

 

 

Get the Most Out Of Your Test Data – MuSigma 2021-02

Speakers: Chris Gross, Gross Consulting

Description:

The data not used in parameterizing your model is very valuable. Unleash the value in the out-of-sample data in both its validation and test roles. Chris will discuss the following approaches:

      -Bootstrap samples to provide statistical tests of the difference between alternative models, without assuming an underlying distribution

-Measurement not only of the fit at a granular level, but also measurement of bias, most easily seen at an aggregated level

-Provide empirical distributions of residuals, capturing process, parameter, and model risk, and reflecting differences in the distribution across different variables. Use bootstrapping techniques to flesh out the distribution.

 

 

Assigning “Minimum Variance” Weights to Reserve Methods – MuSigma 2020-12

Speakers: Chris Gross, Gross Consulting

Description:

When determining how much weight to give various reserve indications, it is useful to use weights that reduce the variance of the resulting weighted reserve estimate.

The variance of the reserve for each method is important for the calculation of the optimal weights, as are the correlations between the methods. The indicated weighting itself can provide useful insight into the reserve analysis.

Chris will discuss this approach, as well as interpretations of various resulting weightings. He will also discuss practical considerations when applying this type of approach.

 

Identically Distributed? Don’t Bet On It! – MuSigma 2020-11

Speakers: Chris Gross, Gross Consulting

Description:

The assumption that observations and model errors are identically distributed is so common in statistical inference that it often is simply overlooked. However, this assumption is invalid in many of the questions that we face as actuaries.

When this assumption is invalid, biased estimates can result, particularly when using multiplicative models.

Chris will give examples of this problem, illustrate methods to test the assumption, and discuss approaches to correcting for the assumption when it is invalid.

Presented by Gross Consulting, the MuSigma Webinar Series is an excellent opportunity for actuaries and insurance professionals to engage in free, interactive, and relevant content designed to provide quality continuing education.

Building a Cloud Data Analytics Platform – MuSigma 2020-10

Speakers: Jeff White and Bret Shroyer, Gross Consulting

Description:

Have you been thinking about modernizing your current analytical data platform? Have you considered using the Cloud? Technology has changed significantly in recent years – if your data tech stack is 5-10+ years old, you are missing out on fantastic new tooling. Because the change in data technology has stabilized in the last year, now is a good time to consider your options.

In this webinar, we will bring you up to speed on the developments in data technology, including the benefits of moving to the cloud. We will also provide overarching principles you should apply to your platform. Last, we will walk you through the process of how to build your platform, with considerations at every step.

Presented by Gross Consulting, the MuSigma Webinar Series is an excellent opportunity for actuaries and insurance professionals to engage in free, interactive, and relevant content designed to provide quality continuing education.