Stephen Richards
Managing Director
Articles written by Stephen Richards
The renewed importance of place
In my previous blog I showed how suddenly the excess deaths rose in Scotland and England & Wales due to the ongoing COVID-19 pandemic. I plotted the excess weekly mortality in two separate graphs because the two countries had such a similar experience that a single figure would have looked muddled.
A week is a long time in a pandemic
According to British Prime Minister Harold Wilson, "a week is a long time in politics". As with politics, so also with the ongoing COVID-19 pandemic.
Significantly enhancing your models
In building a mortality model (or any other kind of risk model) it is usually best to build a single, over-arching model rather than split the data into sub-groups (an approach called stratification, the disadvantages of which are discussed in Macdonald et al (2018)). One obvious reason is to reduce the total number of parameters: why fit two parameters for age when one will do?
COVID-19 mortality and sex
I recently looked at the progression of Covid-19 mortality risk with age. As with all-cause mortality, another risk factor for Covid-19 is biological sex.
COVID-19 mortality and age
When faced with a pandemic disease, such as the ongoing Covid-19 crisis, a multi-layered approach is useful.
Wash your hands, live longer
As the coronavirus circles the globe, the only thing spreading faster is disinformation on so-called "social media". In addition to ridiculous conspiracy theories, quack preventions range from the ineffective to the downright dangerous.
Best practice in mortality work - regulatory comments
In a letter to the Chief Actuaries of UK insurance businesses, Malik (2019) highlighted two aspects of what the regulator regards as good practice in mortality work
Mortality crossover
In a previous blog I discussed the importance of mortality convergence to actuaries, i.e. how mortality differentials narrow with age and how this interacts with discounting of cashflows.
Model selection
When fitting a mortality model, analysts are faced with the decision of which risk factors to include or exclude. One way of doing this is to look for the improvement in an information criterion that balances the fit against the number of parameters. The bigger the improvement in the information criterion, the more strongly the model with the smaller value is preferred.
Matrix repair
When fitting a statistical model we want two things as a minimum: