Huntington's disease,critical illness insurance and life insurance

We describe briefly a model of Huntington's disease (HD), a highly penetrant, dominantly inherited, fatal neurological disorder. Although it is a single-gene disorder, mutations are variable in their effects, depending on the number of times that the CAG trinucleotide is repeated in a certain region of the HD gene. The model covers: (a) rates of onset, depending on CAG repeat length as well as age; (b) post-onset rates of mortality; and (c) the distribution of CAG repeat lengths in the population. Using these, we study the critical illness and life insurance markets. We calculate premiums based on genetic test results that disclose the CAG repeat length, or more simply on a family history of HD. These vary widely with age and policy term; some are exceptionally high, but in a large number of cases cover could be offered within normal underwriting limits. We then consider the possible costs of adverse selection, in terms of increased premiums, under various possible moratoria on the use of genetic information, including family history. These are uniformly very small, because of the rarity of HD, but do show that the costs would be much larger in relative terms if family history could not be used in underwriting. We point out some difficulties involved in applying a moratorium that recognises simply a dichotomy between ‘carriers’ and ‘non-carriers’ of any mutation in a gene when these mutations are, in fact, very variable in their effects. These complexities suggest that restrictions on the disclosure, rather than on the use, of genetic information, if it became established as a principle, could deprive insurers of information needed for risk management even if not used in underwriting.     

The Spatial Durbin Model and the Common Factor Tests

Abstract

The spatial Durbin model occupies an interesting position in the field of spatial econometrics. It is the reduced form of a model with cross-sectional dependence in the errors and it may be used as the nesting equation in a more general approach of model selection. Specifically, in this equation we obtain the common factor tests (of which the likelihood ratio is the best known) whose objective is to discriminate between substantive and residual dependence in an apparently misspecified equation. Our paper tries to delve deeper into the role of the spatial Durbin model in the problem of specifying a spatial econometric model. We include a Monte Carlo study related to the performance of the common factor tests presented in the paper in small sample sizes.     

Local asymptotic normality for randomly censored models with applications to rank tests

For randomly right censored models we study the asymptotic behaviour of linear (rank) statistics under local alternatives. The results can be used to evaluate the asymptotic power of the corresponding tests. For instance we treat the question how to choose the best scores in order to derive asymptotically optimal (rank) tests under certain alternatives.     

Performance hypothesis testing with the Sharpe ratio: The case of hedge funds

As recent research highlights that the Sharpe ratio has a decision theoretic foundation even in the case of asymmetric or fat-tailed excess returns and thus is adequate even for the evaluation of hedge funds, this note provides the first Sharpe ratio based performance analysis of the hedge fund market. Furthermore, it addresses the important practical question whether the choice of hypothesis test used to statistically compare Sharpe ratios can influence an investor’s hedge fund selection process. Our key findings are as follows: (i) Only a small fraction of hedge funds in our large dataset can significantly outperform passive investments in corresponding hedge fund indices. (ii) Especially in the presence of autocorrelated or skewed excess returns, the traditional test of Jobson and Korkie, 1981, Memmel, 2003 tends to overstate the number of significant outperformers and thus provides potentially misleading information for investors. Decision makers are advised to use the bootstrap test of Ledoit and Wolf (2008) allowing robust and more reliable inference.     

Reciprocity,Multiplexity, and Exchange: Measures

We introduce measures and statistical tests for multiplexity and exchange that are modeled on similar ideas developed for reciprocity quite early in the history of social network research. As properties of a multi-relation network, multiplexity, and exchange have almost as ancient a history as reciprocity, but have not been as intensively investigated from a methodological perspective. Multiplexity refers to the extent to which two ties, for example, advice and friendship, coincide over population; that is, do respondents name the same people as friends as the persons they nominate as individuals from who they seek advice. Exchange refers to the extent to which a tie of one type directed from person i to person j is returned by a tie of another type from j to i. We conceive of the current paper as the first part of a two-part paper, wherein the second part explores specific theoretical models for multiplexity and exchange.     

Performance of French intensive care units: A directional distance function approach at the patient level

By using a novel adaptation of the free-disposable hull analysis of productivity, we assess the medical and technical efficiency of patient care in 25 Parisian intensive care units (ICUs) during 2000. The robust free disposable hull (RFDH) as defined by [Cazals et al., 2002. Nonparametric frontier estimation: a robust approach. Journal of Econometrics 106, 1–25] reduces the impact outliers may have on findings by employing Monte-Carlo techniques and repeated sample selection. Among our key findings, there was no overall significant correlation between medical and technical efficiencies for all the ICUs, therefore performing well in one does not guarantee good performance in the other. We also found that over 80% of inefficiency is concentrated in less than 20% of the sampled patients.     

Purchasing Power Parity under Regime Shifts: An Application to Asian Countries*

This paper re‐examines the long‐run purchasing power parity (PPP) relationship for nine Asian countries relative to the USA and Japan during a period containing significant structural breaks. The relevance of considering structural breaks in PPP tests is demonstrated by utilizing the Johansen et al. (2000) procedure that allows for up to two pre‐determined structural breaks. Using conventional tests without considering breaks, one is able to reject the null of no cointegration for only four countries. The Johansen et al. procedure clearly demonstrates the importance of allowing for structural breaks and provides strong support for long‐run PPP for all the countries, regardless of the base country, except in the case of the Philippines vis‐à‐vis Japan. The Hansen–Johansen parameter constancy test indicates stability for all the countries except the Philippines relative to the USA and Malaysia relative to Japan.     

Robust min–max portfolio strategies for rival forecast and risk scenarios

We consider an extension of the Markowitz mean–variance optimization framework to multiple return and risk scenarios. It is well known that asset return forecasts and risk estimates are inherently inaccurate. The method proposed provides a means for considering rival representations of the future. The optimal portfolio is computed, simultaneously with the worst case, to take account of all rival scenarios. This is a min-max strategy which is essentially equivalent to a robust pooling of the scenarios. Robustness is ensured by the noninferiority of min–max. For example, a basic worst-case optimal return is guaranteed in view of multiple return scenarios. If robustness happens to have too high a cost, guided by the min–max pooling, it is also possible to explore other pooling alternatives. A min–max algorithm is used to solve the problem and illustrate the robust character of min–max with return and risk scenarios. We study the properties of the min–max risk–return frontier and compare with the potentially suboptimal worst-case where the investment strategy and the worst case are computed separately.     

VAN DE HULST ON ROBUST STATISTICS: A HISTORICAL NOTE

This paper provides a discussion of an unpublished set of notes written in 1942 by the Dutch astronomer H.C. VAN DE HULST. In these notes VAN DE HULST derives the asymptotic variances of M– estimators as well as trimmed means and concludes that the asymptotic variance of what is now called HUBER'S estimator is the same as that of a trimmed mean. This conclusion is usually ascribed to BICKEL (1965). A letter written by D. VAN DANTZIG in 1943 providing a critical evaluation of. VAN DE HULST'S results, adds interest to this suprisingly early contribution to the theory of robust statistics.     

Diagnostics cannot have much power against general alternatives

Model diagnostics are shown to have little power unless alternative hypotheses can be narrowly defined. For example, the independence of observations cannot be tested against general forms of dependence. Thus, the basic assumptions in regression models cannot be inferred from the data. Equally, the proportionality assumption in proportional-hazards models is not testable. Specification error is a primary source of uncertainty in forecasting, and this uncertainty will be difficult to resolve without external calibration. Model-based causal inference is even more problematic.