PirateGrunt goes to the CLRS

Yesterday, I had the great pleasure to speak about using R for loss reserving at the Casualty Loss Reserving Seminar in Boston. My time was spent talking about MRMR, an R package that I’ve created. Version 0.1.2 is now on CRAN, but as there are a couple of bugs, I’d suggest waiting until version 0.1.3 is released. This should happen in a week or two. Attendees of the workshop got to see a demonstration of version 0.1.3 and there were some great questions and discussion. If you’re interested in having a look at the package, you may find the source and binary on my Github site. I’ll provide a more detailed look at MRMR once version 0.1.3 is up on CRAN.

Immediately before my segment of the workshop, Dan Murphy gave an overview of ChainLadder, the original standard bearer for loss reserving in R. Next door, workshop attendees were learning about how to apply Bayesian MCMC techniques to loss reserving.

This morning, I saw two fantastic presentations from Roger Hayne and Jim Guszcza about stochastic loss reserving. Roger mentioned near the top of his talk that a tool like R now allows actuarial practitioners to carry out highly customizable stochastic models through direct optimization of the likelihood function. I couldn’t agree more.

Am I excited? I’m very excited. There is a lot of very cool stuff happening in loss reserving. R makes most of this advanced modeling possible. Certainly my own appreciation for and understanding of a more statistically grounded approach to actuarial thought has been aided immeasurably by my experience with R.

About these ads

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 264 other followers

%d bloggers like this: