Search

Purpose: Diseases characterized by a continuous trait can be defined by setting a cut-off point for the disease measure in question, accepting some misclassification. The 97.5th percentile is commonly used as a cut-off point. However, it is unclear whether this percentile is the optimal cut-off point from the point of view of risk-factor analysis. The optimal cut-off point for risk-factor analysis can be found with a statistical method that minimizes the effect of misclassification. We applied this method to glaucomatous optic neuropathy. Here, the continuous trait is the cup-disc ratio. The aim of this study was to determine the optimal cup-disc ratio cut-off point for risk-factor analysis in population-based epidemiology.Methods: All participants in the population-based Rotterdam Study underwent intraocular pressure (IOP) measurements, assessment of the cup-disc ratio with the Heidelberg Retina Tomograph (HRT) and visual field testing. In the statistical method, the cup-disc ratio (the continuous trait) and the IOP (a major risk factor) were independent variables and glaucomatous visual field loss (the true glaucoma endpoint) the dependent variable in a logistic regression model. The optimal cup-disc ratio cut-off point was found by minimizing the influence of IOP in this model. Variability of the approach was assessed by using a bootstrap resampling technique.Results: Of 2444 included participants, 93 had glaucomatous visual field loss. The median optimal cup-disc ratio cut-off point was the 97.0th percentile with a 95% central range from 95.5 to 98.5.Conclusion: The optimal cup-disc ratio cut-off point for risk-factor analysis is close to the commonly used 97.5th percentile.

Recently, a prediction rule for developing neurological sequelae after childhood bacterial meningitis was developed on a small derivation set. Before implementing in practice a prediction rule must first be tested in new patients (external validation). Our aim was to study the external validity of this rule and, if necessary, to update the rule. The prediction rule was tested on newly available data (validation set) by assessing the rule's calibration and discrimination. We updated the prediction rule by adding extra predictors and re-estimating the regression coefficients of the original predictors in the combined datasets. The rule showed poor agreement between predicted risks and observed frequencies. The ROC area was 0.65 (95% CI 0.57-0.72), which was statistically significantly lower than in the derivation set (0.87 (0.78-0.96)), p-value

The (Generalized) AutoRegressive Conditional Heteroscedasticity [(G)ARCH] model is tested for daily data on 22 exchange rates and 13 stock market indices using the standard Lagrange Multiplier [LM] test for GARCH and a LM test that is resistant to patches of additive outliers. The data span two samples of five years ranging from 1986 to 1995. Using asymptotic arguments and Monte Carlo simulations, in which the empirical method is evaluated, it is shown that patches of outliers can have significant effects on test outcomes. The main empirical result is that spurious GARCH is found in about 40% of the cases, while in many other cases evidence of GARCH is found even though such sequences of extraordinary observations seem to be present. © 2004 Taylor and Francis Ltd.

Testing mutual independence among several random vectors of arbitrary dimensions is a challenging problem in Statistics, and it has gained considerable interest in recent years. In this article, we propose some nonparametric tests based on different notions of ranks of nearest neighbour. These proposed tests can be conveniently used for high dimensional data, even when the dimensions of the random vectors are larger than the sample size. We investigate the performance of these tests on several simulated and real data sets and also use them in identifying causal relationships among the random vectors. Our numerical results show that they can outperform state-of-the-art tests in a wide variety of examples.