I've recently begun working as a quant for a large bank, and one of my first tasks will be to improve the model determining the risk exposure of their insurance portfolio. The portfolio is fairly diverse, and contains a number of types of insurance policies; life, accident, property, terrorism, auto, mortgage, and a number of others. The model I'm working with is used to estimate economic capital (EC) for the portfolio. Currently, the model is simply the worst-case scenario, which is grossly overestimating the EC required. I'm supposed to improve the model to (more) accurately predict what the exposure risk actually is.
I'm fairly new to the financial world (my background is modeling biological systems), so I'm curious (1) where I should search for published data on modeling insurance risk, and (2) whether anyone here is familiar with any specific approaches with which I should be familiar. I've found the Wharton financial institution, and I found a paper which seems to be a good starting point, but I'm looking for other resources.