Рет қаралды 2,713
Today’s show is a highly practical part 2 (of 3!) to the show I did with Steve and John Major on their book: Pricing Insurance Risk, Theory and Practice and we’ll be walking through a python package Steve built that enables us to work some examples of pricing risk.
This video is literally us walking through a google colab notebook Steve put together and executing real code to explore how to use Steve's package to price insurance. This first part of the walkthrough starts fairly gently with the setup and ends on some basic pricing methods. We'll get into portfolio and capital modeling in part 2.
A link to the colab notebook we're using:
colab.research...
Topics:
Part I: For Reinsurance and Large Account Pricing Actuaries
*Setup and import aggregate components
*Insurance example: commercial auto risk retention group
**Plotting
**Statistics and probability functions
**How does aggregate work?
**Three alternative ways to specify exposure
**Error analysis
*Apply reinsurance
**Plot, accessing underlying density_df dataframes
**Reinsurance impact on capital at different probability thresholds
**Ceded losses
**Layering occurrence losses
**Aggregate cover
**Density functions for gross, net occ, and net from density_df dataframe
**Survival functions for gross, net occ, and net
*Pricing with Distortions
**Price gross with assets at 99 percent level
**Side bar: of bid/ask spread
**Distortion pricing calculation by hand
**Compare gross, ceded, and net pricing
Documentation for the package: aggregate.read...
Source code: github.com/myn...
How-to videos • aggregate
Installation: pypi.org/proje...
Steve Mildenhall is head of analytics at QualRisk, an insurance consulting firm, formerly Assistant Professor of Actuarial Science at St John’s University and before that CEO of Analytics at Aon.
Twitter: @davecwright
Surprise, It's Insurance mailing list: notunreasonabl...
Linkedin: / david-wright-73661214
Social Science of Insurance Essays: notunreasonabl...