Stochastic mortality models are important for a variety of actuarial tasks, from best-estimate forecasting to assessment of risk capital requirements. However, the mortality shock associated with the Covid-19 pandemic of 2020 distorts forecasts by (i) biasing parameter estimates, (ii) biasing starting points, and (iii) inflating variance. Stochastic mortality models therefore require outlier-robust methods for forecasting. Objective methods are required, as outliers are not always obvious on visual inspection. In this paper we look at the robustification of three broad classes of forecast: univariate time indices (such as in the Lee-Carter and APC models); multivariate time indices (such as in the Cairns-Blake-Dowd and newer Tang-Li-Tickle model families); and penalty projections (such as with the 2D P-spline model). In each case we identify outliers using quantitative methods, then co-estimate outlier effects along with other parameters. Doing so removes the bias and distortion to the forecast caused by a mortality shock, while providing a robust starting point for projections. Illustrations are given for various models in common use.

In Ghana, the economic costs of violent conflicts (loss of jobs and revenue, drop in gross domestic product, and disruption of production chain) have been well documented. However, there is little scholarly work on such conflicts’ human security costs, a critical element in contemporary security management arrangements. This study examines the costs of violent conflicts from a human security perspective using the farmer–herder conflict in the Agogo community in the Ashanti region of Ghana as a case study. A total of 400 participants were randomly and purposely selected. A χ2 test was employed to determine the association of rape, armed robbery, arson and murder with the farmer–herder conflict. The study found that at a 5% margin of error, the p value for armed robbery was 0.01, rape was 0.01, arson was 0.03 and murder was 0.00, indicating a close association between these variables and the conflict.

To establish flutter onset boundaries on the flight envelope, it is required to determine the flutter onset dynamic pressure. Proper selection of a flight flutter prediction technique is vital to flutter onset speed prediction. Several methods are available in literature, starting with those based on velocity damping, envelope functions, flutter margin, discrete-time Autoregressive Moving Average (ARMA) modelling, flutterometer and the Houbolt–Rainey algorithm. Each approach has its capabilities and limitations. To choose a robust and efficient flutter prediction technique from among the velocity damping, envelope function, Houbolt–Rainey, flutter margin and auto-regressive techniques, an example problem is chosen for their evaluation. Hence, in this paper, a three-degree-of-freedom model representing the aerodynamics, stiffness and inertia of a typical wing section is used(1). The aerodynamic, stiffness and inertia properties in the example problem are kept the same when each of the above techniques is used to predict the flutter speed of this aeroelastic system. This three-degree-of-freedom model is used to generate data at speeds before initiation of flutter, during flutter and after occurrence of flutter. Using these data, the above-mentioned flutter prediction methods are evaluated and the results are presented.

Advanced machine learning techniques like Gaussian process regression and multi-task learning are novel in the area of wine price prediction; previous research in this area being restricted to parametric linear regression models when predicting wine prices. Using historical price data of the 100 wines in the Liv-Ex 100 index, the main contributions of this paper to the field are, firstly, a clustering of the wines into two distinct clusters based on autocorrelation. Secondly, an implementation of Gaussian process regression on these wines with predictive accuracy surpassing both the trivial and simple ARMA and GARCH time series prediction benchmarks. Lastly, an implementation of an algorithm which performs multi-task feature learning with kernels on the wine returns as an extension to our optimal Gaussian process regression model. Using the optimal covariance kernel from Gaussian process regression, we achieve predictive results which are comparable to that of Gaussian process regression. Altogether, our research suggests that there is potential in using advanced machine learning techniques in wine price prediction. (JEL Classifications: C6, G12)

In this paper, the Auto-Regressive Moving-Averaging (ARMA) neural networks (NNs) will be incorporated for predicting the differential Global Positioning System (DGPS) pseudorange correction (PRC) information. The neural network is employed to realize the time-varying ARMA implementation. Online training for real-time prediction of the PRC enhances the continuity of service on the differential correction signals and therefore improves the positioning accuracy. When the PRC signal is lost, the ARMA neural network predicted PRC would temporarily provide correction data with very good accuracy. Simulation is conducted for evaluating the ARMA NN based DGPS PRC prediction accuracy. A comparative performance study based on two types of ARMA neural networks, i.e. Back-propagation Neural Network (BPNN) and General Regression Neural Network (GRNN), will be provided.