The COVID-19 pandemic requires that actuaries track short-term mortality fluctuations in the portfolios they manage. This demands methods that not only operate over much shorter time periods than a year but that also deal with reporting delays. In this paper, we consider a semi-parametric approach for tracking portfolio mortality levels in continuous time. We identify both seasonal patterns and mortality shocks, thus providing a comparison benchmark for the impact of COVID-19 in terms of a portfolio’s own past experience. A parametric model is presented to allow for the average impact of seasonal variation and also reporting delays. We find that an estimate of mortality reporting delays can be made from a single extract of experience data. This can be used to forecast unreported deaths and improve estimates of recent mortality levels. Results are given for annuity portfolios in France, the UK and the USA.

Nowcasting, that is, forecasting the current economic conditions, is a key ingredient for decision making, but it is complex, even more so for a small open economy, due to the higher volatility of its GDP. In this paper, we review the required steps, taking Luxembourg as an example. We consider both standard and alternative indicators, used as inputs in several nowcasting methods, including various factor and machine learning models. Overall, mixed frequency dynamic factor models and neural networks perform well, both in absolute terms and in relative terms with respect to a benchmark autoregressive model. The gains are larger during problematic times, such as the financial crisis and the recent Covid period.

Expenditure-side and income-side gross domestic product (GDP) are measured at the quarterly frequency and contain measurement error. Econometric methods exist for producing reconciled estimates of underlying true GDP from these noisy estimates. Recently, the authors of this paper developed a mixed-frequency reconciliation model which produces monthly estimates of true GDP. In the present paper, we investigate whether this model continues to work well in the face of the extreme observations that occurred during the pandemic year and consider several extensions of it. These include stochastic volatility and error distributions that are fat-tailed or explicitly allow for outliers.

We propose a nowcasting system to obtain real-time predictive intervals for the first-release of UK quarterly GDP growth that can be implemented in a menu-driven econometric software. We design a bottom-up approach: forecasts for GDP components (from the output and the expenditure approaches) are inputs into the computation of probabilistic forecasts for GDP growth. For each GDP component considered, mixed-data-sampling regressions are applied to extract predictive content from monthly and quarterly indicators. We find that predictions from the nowcasting system are accurate, in particular when nowcasts are computed using monthly indicators 30 days before the GDP release. The system is also able to provide well-calibrated predictive intervals.

For more than four decades, the New Zealand Institute of Economic Research (NZIER) has conducted a two-question, quarterly survey of architect forecasts of public and private sector construction expenditure. This qualitative survey is published one week after the end of each quarter and nine weeks ahead of the official quantitative data thereby giving architect opinion nowcasting status. This paper covers selected aspects of this unexplored series with particular reference to residential housing construction and the value-added information from architects as nowcasters. Specifically, we consider several qualitative-to-quantitative conversion methods, in-sample and out-of-sample performance, cyclical features and respondent dynamics. Although our work relates to architects — a sub-sector of the service industry — our results have a wider application to business survey questions using ordered qualitative responses.

We consider the reasons for nowcasting, the timing of information and sources thereof, especially contemporaneous data, which introduce different aspects compared to forecasting. We allow for the impact of location shifts inducing nowcast failure and nowcasting during breaks, probably with measurement errors. We also apply a variant of the nowcasting strategy proposed in Castle and Hendry (2009) to nowcast Euro Area GDP growth. Models of disaggregate monthly indicators are built by automatic methods, forecasting all variables that are released with a publication lag each period, then testing for shifts in available measures including survey data, switching to robust forecasts of missing series when breaks are detected.

A system has been developed and made operational at the Met Office for the real-time diagnosis of soil state and surface hydrology. It is based on the Met Office Surface Exchanges Scheme (MOSES) modified to take account of unresolved soil and topographic heterogeneity when calculating surface runoff by incorporating a Probability Distributed Moisture (PDM) scheme developed by the Centre for Ecology and Hydrology. The implementation of MOSES-PDM in the Met Office's Nimrod nowcasting system is described. High resolution soil characteristics and land cover data, together with Nimrod's analyses of precipitation amount and type, cloud cover and near-surface atmospheric variables are used to drive MOSES-PDM. Hourly values of snowmelt, runoff, net surface radiation, evaporation, potential evaporation, soil temperature, soil moisture and soil moisture deficit are calculated on a 5 km grid.

A model to deduce cloudiness from automatic measurements of synoptic and road weather stations during winter nights is introduced. The (height-adjusted) cloud amount of each station can be determined from near-surface temperature measurements and precipitation information only. By using the difference between the air temperature at surface level and the air temperature 2 m above ground, a correlation coefficient is achieved of up to 0.91 between the calculated cloud amount and observations. The model is applied to a dense network in central Switzerland in order to obtain a two-dimensional cloud map by interpolation. With the help of a tracking algorithm, the displacement of cloud patterns is nowcasted. The procedure was tested using data from winter 2003–4. It works successfully with a forecast range up to about 90 minutes. The results are used to predict a change in surface temperature (in cases with changing cloudiness), and thus allow nowcasting of slippery roads or ground frost.