Selecting from amongst non–nested conditional variance models: information criteria and portfolio determination

We consider the finite sample properties of model selection by information criteria in conditionally heteroscedastic models. Recent theoretical results show that certain popular criteria are consistent in that they will select the true model asymptotically with probability 1. To examine the empirical relevance of this property, Monte Carlo simulations are conducted for a set of non–nested data generating processes (DGPs) with the set of candidate models consisting of all types of model used as DGPs. In addition, not only is the best model considered but also those with similar values of the information criterion, called close competitors, thus forming a portfolio of eligible models. To supplement the simulations, the criteria are applied to a set of economic and financial series. In the simulations, the criteria are largely ineffective at identifying the correct model, either as best or a close competitor, the parsimonious GARCH(1, 1) model being preferred for most DGPs. In contrast, asymmetric models are generally selected to represent actual data. This leads to the conjecture that the properties of parameterizations of processes commonly used to model heteroscedastic data are more similar than may be imagined and that more attention needs to be paid to the behaviour of the standardized disturbances of such models, both in simulation exercises and in empirical modelling.

The conditional performance of UK property funds

The evaluation of investment fund performance has been one of the main developments of modern portfolio theory. Most studies employ the technique developed by Jensen (1968) that compares a particular fund's returns to a benchmark portfolio of equal risk. However, the standard measures of fund manager performance are known to suffer from a number of problems in practice. In particular previous studies implicitly assume that the risk level of the portfolio is stationary through the evaluation period. That is unconditional measures of performance do not account for the fact that risk and expected returns may vary with the state of the economy. Therefore many of the problems encountered in previous performance studies reflect the inability of traditional measures to handle the dynamic behaviour of returns. As a consequence Ferson and Schadt (1996) suggest an approach to performance evaluation called conditional performance evaluation which is designed to address this problem. This paper utilises such a conditional measure of performance on a sample of 27 UK property funds, over the period 1987-1998. The results of which suggest that once the time varying nature of the funds beta is corrected for, by the addition of the market indicators, the average fund performance show an improvement over that of the traditional methods of analysis.

Comment on ``Conditional Exceedance Probabilities''

In a recent paper, Mason et al. propose a reliability test of ensemble forecasts for a continuous, scalar verification. As noted in the paper, the test relies on a very specific interpretation of ensembles, namely, that the ensemble members represent quantiles of some underlying distribution. This quantile interpretation is not the only interpretation of ensembles, another popular one being the Monte Carlo interpretation. Mason et al. suggest estimating the quantiles in this situation; however, this approach is fundamentally flawed. Errors in the quantile estimates are not independent of the exceedance events, and consequently the conditional exceedance probabilities (CEP) curves are not constant, which is a fundamental assumption of the test. The test would reject reliable forecasts with probability much higher than the test size.

Conditional transgenic expression of fibroblast growth factor 9 in the adult mouse heart reduces heart failure mortality after myocardial infarction

BACKGROUND: Fibroblast growth factor 9 (FGF9) is secreted from bone marrow cells, which have been shown to improve systolic function after myocardial infarction (MI) in a clinical trial. FGF9 promotes cardiac vascularization during embryonic development but is only weakly expressed in the adult heart. METHODS AND RESULTS: We used a tetracycline-responsive binary transgene system based on the α-myosin heavy chain promoter to test whether conditional expression of FGF9 in the adult myocardium supports adaptation after MI. In sham-operated mice, transgenic FGF9 stimulated left ventricular hypertrophy with microvessel expansion and preserved systolic and diastolic function. After coronary artery ligation, transgenic FGF9 enhanced hypertrophy of the noninfarcted left ventricular myocardium with increased microvessel density, reduced interstitial fibrosis, attenuated fetal gene expression, and improved systolic function. Heart failure mortality after MI was markedly reduced by transgenic FGF9, whereas rupture rates were not affected. Adenoviral FGF9 gene transfer after MI similarly promoted left ventricular hypertrophy with improved systolic function and reduced heart failure mortality. Mechanistically, FGF9 stimulated proliferation and network formation of endothelial cells but induced no direct hypertrophic effects in neonatal or adult rat cardiomyocytes in vitro. FGF9-stimulated endothelial cell supernatants, however, induced cardiomyocyte hypertrophy via paracrine release of bone morphogenetic protein 6. In accord with this observation, expression of bone morphogenetic protein 6 and phosphorylation of its downstream targets SMAD1/5 were increased in the myocardium of FGF9 transgenic mice. CONCLUSIONS: Conditional expression of FGF9 promotes myocardial vascularization and hypertrophy with enhanced systolic function and reduced heart failure mortality after MI. These observations suggest a previously unrecognized therapeutic potential for FGF9 after MI.

Conditional mean functions of non-linear models of US output

We compare a number of models of post War US output growth in terms of the degree and pattern of non-linearity they impart to the conditional mean, where we condition on either the previous period's growth rate, or the previous two periods' growth rates. The conditional means are estimated non-parametrically using a nearest-neighbour technique on data simulated from the models. In this way, we condense the complex, dynamic, responses that may be present in to graphical displays of the implied conditional mean.

Forecasting exchange rate volatility using conditional variance models selected by information criteria

This paper uses appropriately modified information criteria to select models from the GARCH family, which are subsequently used for predicting US dollar exchange rate return volatility. The out of sample forecast accuracy of models chosen in this manner compares favourably on mean absolute error grounds, although less favourably on mean squared error grounds, with those generated by the commonly used GARCH(1, 1) model. An examination of the orders of models selected by the criteria reveals that (1, 1) models are typically selected less than 20% of the time.

Conditional male dimorphism and alternative reproductive tactics in a Neotropical arachnid (Opiliones)

In arthropods, most cases of morphological dimorphism within males are the result of a conditional evolutionarily stable strategy (ESS) with status-dependent tactics. In conditionally male-dimorphic species, the status` distributions of male morphs often overlap, and the environmentally cued threshold model (ET) states that the degree of overlap depends on the genetic variation in the distribution of the switchpoints that determine which morph is expressed in each value of status. Here we describe male dimorphism and alternative mating behaviors in the harvestman Serracutisoma proximum. Majors express elongated second legs and use them in territorial fights; minors possess short second legs and do not fight, but rather sneak into majors` territories and copulate with egg-guarding females. The static allometry of second legs reveals that major phenotype expression depends on body size (status), and that the switchpoint underlying the dimorphism presents a large amount of genetic variation in the population, which probably results from weak selective pressure on this trait. With a mark-recapture study, we show that major phenotype expression does not result in survival costs, which is consistent with our hypothesis that there is weak selection on the switchpoint. Finally, we demonstrate that switchpoint is independent of status distribution. In conclusion, our data support the ET model prediction that the genetic correlation between status and switchpoint is low, allowing the status distribution to evolve or to fluctuate seasonally, without any effect on the position of the mean switchpoint.

Regression Models with Heteroscedasticity using Bayesian Approach

In this paper, we compare the performance of two statistical approaches for the analysis of data obtained from the social research area. In the first approach, we use normal models with joint regression modelling for the mean and for the variance heterogeneity. In the second approach, we use hierarchical models. In the first case, individual and social variables are included in the regression modelling for the mean and for the variance, as explanatory variables, while in the second case, the variance at level 1 of the hierarchical model depends on the individuals (age of the individuals), and in the level 2 of the hierarchical model, the variance is assumed to change according to socioeconomic stratum. Applying these methodologies, we analyze a Colombian tallness data set to find differences that can be explained by socioeconomic conditions. We also present some theoretical and empirical results concerning the two models. From this comparative study, we conclude that it is better to jointly modelling the mean and variance heterogeneity in all cases. We also observe that the convergence of the Gibbs sampling chain used in the Markov Chain Monte Carlo method for the jointly modeling the mean and variance heterogeneity is quickly achieved.

Genetic Heteroscedasticity for Domestic Animal Traits

Animal traits differ not only in mean, but also in variation around the mean. For instance, one sire’s daughter group may be very homogeneous, while another sire’s daughters are much more heterogeneous in performance. The difference in residual variance can partially be explained by genetic differences. Models for such genetic heterogeneity of environmental variance include genetic effects for the mean and residual variance, and a correlation between the genetic effects for the mean and residual variance to measure how the residual variance might vary with the mean. The aim of this thesis was to develop a method based on double hierarchical generalized linear models for estimating genetic heteroscedasticity, and to apply it on four traits in two domestic animal species; teat count and litter size in pigs, and milk production and somatic cell count in dairy cows. The method developed is fast and has been implemented in software that is widely used in animal breeding, which makes it convenient to use. It is based on an approximation of double hierarchical generalized linear models by normal distributions. When having repeated observations on individuals or genetic groups, the estimates were found to be unbiased. For the traits studied, the estimated heritability values for the mean and the residual variance, and the genetic coefficients of variation, were found in the usual ranges reported. The genetic correlation between mean and residual variance was estimated for the pig traits only, and was found to be favorable for litter size, but unfavorable for teat count.

Informal sanctions and conditional cooperation: : A natural experiment on voluntary contributions to a public good.

In a natural experiment, this paper studies the impact of an informal sanctioning mechanism on individuals’ voluntary contribution to a public good. Cross-country skiers’ actual cash contributions in two ski resorts, one with and one without an informal sanctioning system, are used. I find the contributing share to be higher in the informal sanctioning system (79 percent) than in the non-sanctioning system (36 percent). Previous studies in one-shot public good situations have found an increasing conditional contribution (CC) function, i.e. the relationship between expected average contributions of other group members and the individual’s own contribution. In contrast, the present results suggest that the CC-function in the non-sanctioning system is non-increasing at high perceived levels of others’ contribution. This relationship deserves further testing in lab.

Fitting conditional and simultaneous autoregressive spatial models in hglm

We present a new version (> 2.0) of the hglm package for fitting hierarchical generalized linear models (HGLMs) with spatially correlated random effects. CAR() and SAR() families for conditional and simultaneous autoregressive random effects were implemented. Eigen decomposition of the matrix describing the spatial structure (e.g., the neighborhood matrix) was used to transform the CAR/SAR random effects into an independent, but eteroscedastic, Gaussian random effect. A linear predictor is fitted for the random effect variance to estimate the parameters in the CAR and SAR models. This gives a computationally efficient algorithm for moderately sized problems.