Stephen Richards Profile Picture

Stephen Richards

Managing Director

Articles written by Stephen Richards

The name of the game

We have written frequently on the importance of deduplication for mortality modelling.  In a mortality- or longevity-related transaction, it is critical that the risk-taker performs deduplication when fitting a statistical model to experience data.
Tags: Filter information matrix by tag: deduplication, Filter information matrix by tag: names, Filter information matrix by tag: National Insurance numbers, Filter information matrix by tag: proportion married

A chill wind

In a previous blogs I have looked at seasonal fluctuations in mortality, usually with lower mortality in summer and higher mortality in winter.  The subject of excess winter deaths is back in the news, as the UK experienced heavy mortality in the winter of 2014/15, as demonstrated in Figure 1.

Tags: Filter information matrix by tag: season, Filter information matrix by tag: influenza, Filter information matrix by tag: winter, Filter information matrix by tag: frailty, Filter information matrix by tag: mortality plasticity

What — and when — is a 1:200 event?

The concept of a "one in two hundred" (1:200) event over a one-year time horizon is well established as a reserving standard for insurance in several territories: the ICA in the United Kingdom, the SST in Switzerland and the forthcoming Solvency II standard for the entire European Union. 
Tags: Filter information matrix by tag: Spanish influenza pandemic, Filter information matrix by tag: mortality shocks, Filter information matrix by tag: longevity shocks, Filter information matrix by tag: Solvency II, Filter information matrix by tag: ICA, Filter information matrix by tag: SST, Filter information matrix by tag: VaR, Filter information matrix by tag: value-at-risk

Reviewing forecasts

When making projections and forecasts, it can be instructive to compare them with what actually happened. In December 2002 the CMI published projections of mortality improvements that incorporated the so-called "cohort effect" (CMIB, 2002). These projections were in use by life offices and pension schemes in the United Kingdom from 2003 onwards.

Tags: Filter information matrix by tag: cohort effect, Filter information matrix by tag: mortality projections, Filter information matrix by tag: mortality improvements

Conditional tail expectations

In a recent posting I looked at the calculation of percentiles and quantiles, which underpin many calculations for ICA and Solvency II. Simply put, an \(\alpha\)-quantile is the value which is not expected to be exceeded \(\alpha\times 100\)% of the time. This value is denoted \(Q_{\alpha}\). Mathematically, for a continuous random variable, \(X\), and a given probability level \(\alpha\) we have:

$$\Pr(X\leq Q_\alpha)=\alpha$$

Tags: Filter information matrix by tag: conditional tail expectation, Filter information matrix by tag: quantile, Filter information matrix by tag: percentile, Filter information matrix by tag: coherence, Filter information matrix by tag: subadditivity

Quantiles and percentiles

Quantiles are points taken at regular intervals from the cumulative distribution function of a random variable. They are generally described as q-quantiles, where q specifies the number of intervals which are separated by q−1 points.
Tags: Filter information matrix by tag: quantile, Filter information matrix by tag: percentile, Filter information matrix by tag: Solvency II, Filter information matrix by tag: Excel, Filter information matrix by tag: R language

Creative thinking around longevity risk

The U.K. has been a hotbed of innovation when dealing with the longevity risk found in pension schemes.
Tags: Filter information matrix by tag: longevity risk, Filter information matrix by tag: regulation, Filter information matrix by tag: longevity swaps

Excel's limits

We have written in the past about some of the reasons why we don't use Excel to fit our models.  However, we do use Excel for validation purposes — fitting models using two entirely separate tools is a good way of checking production code.  That said, there are some important limits to Excel, especially when it comes to fitting projection models.
Tags: Filter information matrix by tag: Excel, Filter information matrix by tag: Lee-Carter, Filter information matrix by tag: APC, Filter information matrix by tag: Cairns-Blake-Dowd