In my previous blog I described a real case where so-called artificial intelligence (AI) would have struggled to spot data problems that a (suspicious) human could find.  But what if the input data are clean and reliable?

The short answer for mortality work is that your Poisson model is never truly Poisson. The longer answer is that the true distribution has a similar likelihood, so you will get the same answer from treating it like Poisson.  Your model is pseudo-Poisson, but not actually Poisson.

Last week I presented at the Longevity 18 conference.  My topic was on robustifying stochastic mortality models when the calibrating data contain outliers, such as caused by the COVID-19 pandemic.  A copy of the presentation can be downloaded here, which is based on a paper to be presented at an IFoA sessional meeting in N

The Institute and Faculty of Actuaries (IFoA) is hosting a meeting on robust mortality forecasting in the presence of outliers on 27th November 2023.  A copy of the draft paper and booking details for attendance can be found on the IFoA website.

2.8.7 update

• A new resource area Training tab including datasets and exercises.
• Availability of 2D Period shocks model as per Kirkby & Currie (2010).
• Availability of a CBD M9 model with smoothing on main age term.
• New plot shows the inverse roots of the characteristic polynomials of the autoregressive (AR) and moving-average (MA) parts of the ARIMA model.
• License configurable SAML-based single-sign-on authentication.
• Platform, security and administrative support updates.
• Various fix

in Kleinow & Richards (2016, Table 5) we noted a seeming conundrum: the best-fitting ARIMA model for the time index in a Lee-Carter model also produced much higher value-at-risk (VaR) capital requirements for longevity trend risk.  How could this be?

The late Iain Currie was a long-time advocate of smoothing certain parameters in mortality models.  In an earlier blog he showed how smoothing parameters in the Lee-Carter model could improve the quality of the forecast.  As Iain himself wrote, "this idea is not new" and traced its origins to Delwarde, Denuit & Eilers (2007).

The covid-19 pandemic caused mortality shocks in many countries, and these shocks severely impact the standard forecasting models used by actuaries.  I previously showed how to robustify time-series models with a univariate index (Lee-Carter, APC) and those with a multivariate index (Cairns-Blake-Dowd, Ta

In earlier blogs I discussed two techniques for handling outliers in mortality forecasting models:

• Chen & Liu (1993) for ARIMA models for univariate time series, and