In this book, Andrew Harvey sets out to provide a unified and comprehensive theory of structural time series models. Unlike the traditional ARIMA models, structural time series models consist explicitly of unobserved components, such as trends and seasonals, which have a direct interpretation. As a result the model selection methodology associated with structural models is much closer to econometric methodology. The link with econometrics is made even closer by the natural way in which the models can be extended to include explanatory variables and to cope with multivariate time series. From the technical point of view, state space models and the Kalman filter play a key role in the statistical treatment of structural time series models. The book includes a detailed treatment of the Kalman filter. This technique was originally developed in control engineering, but is becoming increasingly important in fields such as economics and operations research. This book is concerned primarily with modelling economic and social time series, and with addressing the special problems which the treatment of such series poses. The properties of the models and the methodological techniques used to select them are illustrated with various applications. These range from the modellling of trends and cycles in US macroeconomic time series to to an evaluation of the effects of seat belt legislation in the UK.

Forecasting, Structural Time Series Models and the Kalman Filter

Forecasting with artificial neural networks: the state of the art

Background In the face of rapidly changing data, a range of case fatality ratio estimates for coronavirus disease 2019 (COVID-19) have been produced that differ substantially in magnitude. We aimed to provide robust estimates, accounting for censoring and ascertainment biases. Methods We collected individual-case data for patients who died from COVID-19 in Hubei, mainland China (reported by national and provincial health commissions to Feb 8, 2020), and for cases outside of mainland China (from government or ministry of health websites and media reports for 37 countries, as well as Hong Kong and Macau, until Feb 25, 2020). These individual-case data were used to estimate the time between onset of symptoms and outcome (death or discharge from hospital). We next obtained age-stratified estimates of the case fatality ratio by relating the aggregate distribution of cases to the observed cumulative deaths in China, assuming a constant attack rate by age and adjusting for demography and age-based and location-based under-ascertainment. We also estimated the case fatality ratio from individual line-list data on 1334 cases identified outside of mainland China. Using data on the prevalence of PCR-confirmed cases in international residents repatriated from China, we obtained age-stratified estimates of the infection fatality ratio. Furthermore, data on age-stratified severity in a subset of 3665 cases from China were used to estimate the proportion of infected individuals who are likely to require hospitalisation. Findings Using data on 24 deaths that occurred in mainland China and 165 recoveries outside of China, we estimated the mean duration from onset of symptoms to death to be 17·8 days (95% credible interval [CrI] 16·9-19·2) and to hospital discharge to be 24·7 days (22·9-28·1). In all laboratory confirmed and clinically diagnosed cases from mainland China (n=70 117), we estimated a crude case fatality ratio (adjusted for censoring) of 3·67% (95% CrI 3·56-3·80). However, after further adjusting for demography and under-ascertainment, we obtained a best estimate of the case fatality ratio in China of 1·38% (1·23-1·53), with substantially higher ratios in older age groups (0·32% [0·27-0·38] in those aged Interpretation These early estimates give an indication of the fatality ratio across the spectrum of COVID-19 disease and show a strong age gradient in risk of death. Funding UK Medical Research Council.

Estimates of the severity of coronavirus disease 2019: a model-based analysis.

A novel influenza A (H1N1) virus has spread rapidly across the globe. Judging its pandemic potential is difficult with limited data, but nevertheless essential to inform appropriate health responses. By analyzing the outbreak in Mexico, early data on international spread, and viral genetic diversity, we make an early assessment of transmissibility and severity. Our estimates suggest that 23,000 (range 6000 to 32,000) individuals had been infected in Mexico by late April, giving an estimated case fatality ratio (CFR) of 0.4% (range: 0.3 to 1.8%) based on confirmed and suspected deaths reported to that time. In a community outbreak in the small community of La Gloria, Veracruz, no deaths were attributed to infection, giving an upper 95% bound on CFR of 0.6%. Thus, although substantial uncertainty remains, clinical severity appears less than that seen in the 1918 influenza pandemic but comparable with that seen in the 1957 pandemic. Clinical attack rates in children in La Gloria were twice that in adults ( /=15 years: 29%). Three different epidemiological analyses gave basic reproduction number (R0) estimates in the range of 1.4 to 1.6, whereas a genetic analysis gave a central estimate of 1.2. This range of values is consistent with 14 to 73 generations of human-to-human transmission having occurred in Mexico to late April. Transmissibility is therefore substantially higher than that of seasonal flu, and comparable with lower estimates of R0 obtained from previous influenza pandemics.

/pdf/pandemic-potential-of-a-strain-of-influenza-a-h1n1-early-2igit7d602.pdf

Pandemic Potential of a Strain of Influenza A (H1N1) : Early Findings

SUMMARY We study a general univariate smooth transition autoregressive, STAR, model. It contains as a special case the self-exciting threshold autoregressive, SETAR, model. We present three tests for testing linearity against STAR models and discuss their properties. The power of the tests in small samples is investigated by simulation when the alternative is the logistic STAR model. One of the tests is identical to Tsay's (1986) test statistic and is recommended only in a special case. Of the two remaining tests with wider applicability, one seems superior to the other in small samples. It is also more powerful than the CUSUM test recently proposed for testing linearity against SETAR models.

Testing linearity against smooth transition autoregressive models

During the course of an epidemic of a potentially fatal disease, it is important that the case fatality ratio be well estimated. The authors propose a novel method for doing so based on the Kaplan-Meier survival procedure, jointly considering two outcomes (death and recovery), and evaluate its performance by using data from the 2003 epidemic of severe acute respiratory syndrome in Hong Kong, People's Republic of China. They compare this estimate obtained at various points in the epidemic with the case fatality ratio eventually observed; with two commonly quoted, naive estimates derived from cumulative incidence and mortality statistics at single time points; and with estimates in which a parametric mixture model is used. They demonstrate the importance of patient characteristics regarding outcome by analyzing subgroups defined by age at admission to the hospital.

Methods for estimating the case fatality ratio for a novel, emerging infectious disease.

We improve the time complexities for solving the polygonal curve approximation problems formulated by Imai and Iri. The time complexity for approximating any polygonal curve of n vertices with minimum number of line segments can be improved from O(n2log n) to O(n2). The time complexity for approximating any polygonal curve with minimum error can also be improved from O(n2log2n) to O(n2log n). We further show that if the curve to be approximated forms part of a convex polygon, the two problems can be solved in O(n) and O(n2) time respectively for both open and closed polygonal curves.

/pdf/approximation-of-polygonal-curves-with-minimum-number-of-1gbihw7a23.pdf

Approximation of polygonal curves with minimum number of line segments or minimum error

The ecology of health care in Hong Kong.

Response-adaptive designs have been extensively studied and used in clinical trials. However, there is a lack of a comprehensive study of response-adaptive designs that include covariates, despite their importance in clinical trials. Because the allocation scheme and the estimation of parameters are affected by both the responses and the covariates, covariate-adjusted response-adaptive (CARA) designs are very complex to formulate. In this paper, we overcome the technical hurdles and lay out a framework for general CARA designs for the allocation of subjects to K (≥ 2) treatments. The asymptotic properties are studied under certain widely satisfied conditions. The proposed CARA designs can be applied to generalized linear models. Two important special cases, the linear model and the logistic regression model, are considered in detail.

/pdf/asymptotic-properties-of-covariate-adjusted-response-2gee7rj48c.pdf

Asymptotic properties of covariate-adjusted response-adaptive designs

Ruin theory under multi-dimensional risk models is very complex. Even in the two-dimensional case, the problem is challenging. In this paper, we consider a bivariate risk model. Three different types of ruin probabilities are defined. Using some results of one-dimensional risk processes, simple bounds for the two-dimensional ruin probabilities are obtained. Numerical examples and simulation experiments are given to illustrate the tightness of the bounds. A partial integral–differential equation satisfied by the two-dimensional ruin probabilities is derived. Although special cases and examples in this paper provide some exciting results, the problem of ruin probability in a multi-dimensional risk model is still far from solved. We hope that this paper stimulates more research by actuaries in this area.

Wai-Sum Chan

Papers

Methods for estimating the case fatality ratio for a novel, emerging infectious disease.

Approximation of polygonal curves with minimum number of line segments or minimum error

The ecology of health care in Hong Kong.

Asymptotic properties of covariate-adjusted response-adaptive designs

Some results on ruin probabilities in a two-dimensional risk model