Presentations that I hope you find of interest. Typically the length of the presentations are 15-20 minutes unless specified but are here as a longer read than some of the blog posts. Please contact me if you have any queries about the work. The pdf icon should bring up each of the relevant presentations, or there is a video embedded.
Cat Risk and Climate Change: Observations, Models, Theory: - ISCM Meeting, 14th August 2020
The International Society of Catastrophe Managers invited me to speak, along with speakers from Ariel, RMS, AIR, AXIS, JBA Risk and Chubb on the intersection of science and catastrophe modelling. This talk focuses on the evidence for climate change and how robust it is, and we might go about making decisions on how to use the data.
It also talks about how we need to understand small shifts in hazard as they may have consequences for a) trends that might not be spottable in a single historical dataset and b) how they could be influencing the tail.
Efficacy of Hindcasts at Forecasting Insurance Losses - ECMWF UEF2020 Conference, 3rd June 2020
I was lucky enough to have my abstract accepted to talk at the ECMWF "Using ECMWF's Forecasts" presentation. This talk has been an ongoing piece of research to ascertain the lead time with which we can provide useful loss guidance.
I have converted the 1993-2016 winter ensemble forecasts (25 per winter) into a simple loss value, which I've compared against the ECMWF ERA-5 reanalysis, which I use as the ground truth.
The results aren't particularly promising as yet, but this is a still very much a work-in-progress.
Historical Data and Climate Change - EGU, 7th May 2020
This was an invited presentation in a session developed by the Potsdam Institute for Climate Impact Research and Guy Carpenter.
I wanted to use to the talk to present to "good and bad" about using historical data in catastrophe modelling to set rates and also the upsides and downsides of using climate model data: it certainly not a panacea but potentially there is use given that historical data only grows slowly.
I don't think there's a right answer here but the talk was there to get to the key points I think are worth talking about.