Some hypothesis tests for the covariance matrix when the dimension is large compared to the sample size

This paper analyzes whether standard covariance matrix tests work whendimensionality is large, and in particular larger than sample size. Inthe latter case, the singularity of the sample covariance matrix makeslikelihood ratio tests degenerate, but other tests based on quadraticforms of sample covariance matrix eigenvalues remain well-defined. Westudy the consistency property and limiting distribution of these testsas dimensionality and sample size go to infinity together, with theirratio converging to a finite non-zero limit. We find that the existingtest for sphericity is robust against high dimensionality, but not thetest for equality of the covariance matrix to a given matrix. For thelatter test, we develop a new correction to the existing test statisticthat makes it robust against high dimensionality.

How to screen for problematic cannabis use in population surveys: an evaluation of the Cannabis Use Disorders Identification Test (CUDIT) in a Swiss sample of adolescents and young adults.

BACKGROUND/AIMS: Cannabis use is a growing challenge for public health, calling for adequate instruments to identify problematic consumption patterns. The Cannabis Use Disorders Identification Test (CUDIT) is a 10-item questionnaire used for screening cannabis abuse and dependency. The present study evaluated that screening instrument. METHODS: In a representative population sample of 5,025 Swiss adolescents and young adults, 593 current cannabis users replied to the CUDIT. Internal consistency was examined by means of Cronbach's alpha and confirmatory factor analysis. In addition, the CUDIT was compared to accepted concepts of problematic cannabis use (e.g. using cannabis and driving). ROC analyses were used to test the CUDIT's discriminative ability and to determine an appropriate cut-off. RESULTS: Two items ('injuries' and 'hours being stoned') had loadings below 0.5 on the unidimensional construct and correlated lower than 0.4 with the total CUDIT score. All concepts of problematic cannabis use were related to CUDIT scores. An ideal cut-off between six and eight points was found. CONCLUSIONS: Although the CUDIT seems to be a promising instrument to identify problematic cannabis use, there is a need to revise some of its items.

First consumption ever of multiple substances: applying an expert-based taxonomy to a Swiss national sample of adolescents.

The use of multiple legal and illegal substances by adolescents is a growing concern in all countries, but since no consensus about a taxonomy did emerge yet, it is difficult to understand the different patterns of consumption and to implement tailored prevention and treatment programs directed towards specific subgroups of the adolescent population. Using data from a Swiss survey on adolescent health, we analyzed the age at which ten legal and illegal substances were consumed for the first time ever by applying a method combining the strength of both automatic clustering and use of substance experts. Results were then compared to 30 socio-economic factors to establish the usefulness of and to validate our taxonomy. We also analyzed the succession of substance first use for each group. The final taxonomy consists of eight groups ranging from non-consumers to heavy drug addicts. All but four socio-economic factors were significantly associated with the taxonomy, the strongest associations being observed with health, behavior, and sexuality factors. Numerous factors influence adolescents in their decision to first try substances or to use them on a regular basis, and no factor alone can be considered as an absolute marker of problematic behavior regarding substance use. Different processes of experimentation with substances are associated with different behaviors, therefore focusing on only one substance or only one factor is not efficient. Prevention and treatment programs can then be tailored to address specific issues related to different youth subgroups.

Rapid sample pre-treatment prior to GC-MS and GC-MS/MS urinary toxicological screening.

Drug screening is an important issue in clinical and forensic toxicology. Gas chromatography coupled to mass spectrometry (GC-MS) remains the gold standard technique for the screening of unknown compounds in urine samples. However, this technique requires substantial sample preparation, which is time consuming. Moreover, some common drugs such as cannabis cannot be easily detected in urine using general procedures. In this work, a sample preparation protocol for treating 200 μL of urine in less than 30 min is described. The enzymatic hydrolysis of glucuro-conjugates was performed in 5 min thanks to the use of microwaves. The use of a deconvolution software allowed reducing the GC-MS run to 10 min, without impairing the quality of the compound identifications. Comparing the results from 139 authentic urine samples to those obtained using the current routine analysis indicated this method performed well. Moreover, additional 5-min GC-MS/MS programs are described, enabling a very sensitive target screening of 54 drugs, including THC-COOH or buprenorphine, without further sample preparation. These methods appeared as an interesting alternative to immuno-assays based screening. The analytical strategy presented in this article proved to be a promising approach for systematic toxicological analysis (STA) of drugs in urine.

Psychometric evaluation of the IWQOL-Lite (Spanish version) when applied to a sample of obese patients awaiting bariatric surgery

Background Obesity may have an impact on key aspects of health-related quality of life (HRQOL). In this context, the Impact of Weight Quality of Life (IWQOL) questionnaire was the first scale designed to assess HRQOL. The aim of the present study was twofold: to assess HRQOL in a sample of Spanish patients awaiting bariatric surgery and to determine the psychometric properties of the IWQOL-Lite and its sensitivity to detect differences in HRQOL across groups. Methods Participants were 109 obese adult patients (BMI¿ 35 kg/m2) from Barcelona, to whom the following measurement instruments were applied: IWQOL-Lite, Depression Anxiety Stress Scales, Brief Symptom Inventory, and self-perception items. Results Descriptive data regarding the IWQOL-Lite scores obtained by these patients are reported. Principal components analysis revealed a five-factor model accounting for 72.05% of the total variance, with factor loadings being adequate for all items. Corrected itemtotal correlations were acceptable for all items. Cronbach"s alpha coefficients were excellent both for the subscales (0.880.93) and the total scale (0.95). The relationship between the IWQOLLite and other variables supports the construct validity of the scale. Finally, sensitivity analysis revealed large effect sizes when comparing scores obtained by extreme BMI groups. Conclusions This is the first study to report the application of the IWQOL-Lite to a sample of Spanish patients awaiting bariatric surgery and to confirm that the Spanish version of the instrument has adequate psychometric properties.

Out-of-sample forecast tests robust to the choice of window size

This paper proposes new methodologies for evaluating out-of-sample forecastingperformance that are robust to the choice of the estimation window size. The methodologies involve evaluating the predictive ability of forecasting models over a wide rangeof window sizes. We show that the tests proposed in the literature may lack the powerto detect predictive ability and might be subject to data snooping across differentwindow sizes if used repeatedly. An empirical application shows the usefulness of themethodologies for evaluating exchange rate models' forecasting ability.

Equilibrium curves and growth models to deal with forests in transition to uneven-aged structure : application in two sample stands

Abstract

Posttraumatic stress avoidance symptoms as mediators in the development of alcohol use disorders after exposure to childhood sexual abuse in a Swiss community sample.

This study examined the role of posttraumatic stress disorder (PTSD) symptoms of re-experience, avoidance, and hyperarousal in the relationship between different types of trauma and alcohol use disorders (AUD). We used data from 731 trauma-exposed individuals who participated in the first wave of the PsyCoLaus-study. Trauma characteristics were assessed relatively to the occurrence of lifetime PTSD symptoms and AUD. The results suggest that lifetime and childhood sexual abuse as well as overall childhood trauma were directly linked to AUD and PTSD symptoms, in particular to avoidance symptoms. From single symptom clusters PTSD avoidance was found to specifically mediate the trauma-AUD pathway. Both childhood and sexual trauma strongly contribute to the comorbidity of PTSD and AUD and avoidance-type symptoms appear to play a central role in maintaining this association. Hence, the alleviation of avoidance symptoms might be an important target for therapeutic intervention among victims of sexual abuse before specific addiction treatment is initiated.

Posttraumatic stress avoidance symptoms as mediators in the development of alcohol use disorders after exposure to childhood sexual abuse in a Swiss community sample.

This study examined the role of posttraumatic stress disorder (PTSD) symptoms of re-experience, avoidance, and hyperarousal in the relationship between different types of trauma and alcohol use disorders (AUD). We used data from 731 trauma-exposed individuals who participated in the first wave of the PsyCoLaus-study. Trauma characteristics were assessed relatively to the occurrence of lifetime PTSD symptoms and AUD. The results suggest that lifetime and childhood sexual abuse as well as overall childhood trauma were directly linked to AUD and PTSD symptoms, in particular to avoidance symptoms. From single symptom clusters PTSD avoidance was found to specifically mediate the trauma-AUD pathway. Both childhood and sexual trauma strongly contribute to the comorbidity of PTSD and AUD and avoidance-type symptoms appear to play a central role in maintaining this association. Hence, the alleviation of avoidance symptoms might be an important target for therapeutic intervention among victims of sexual abuse before specific addiction treatment is initiated.

Urinary 2,5-hexanedione in workers exposed to n-hexane: influence of the sample treatment

The aim of the present study was to determine 2,5-hexanedione (2,5-HD), a metabolite of n-hexane, by gas chromatography/flame ionization detection in 31 workers exposed to n-hexane after two types of sample pretreatment, i.e., with (total 2,5-HD) and without (free 2,5-HD) acid hydrolysis. The mean urinary 2,5-HD was 0.52 mg/L (free) and 2.88 mg/L (total), this difference being significant (Student t-test, p < 0.05). The differences in the results according to the sample treatment support the need to modify the current Brazilian legislation, which proposes the analysis of 2,5-HD without indicating whether it is the free or total metabolite.

Familial predisposition to hypertension and the association between urinary sodium excretion and blood pressure in a population-based sample of young adults

The reasons for the inconsistent association between salt consumption and blood pressure levels observed in within-society surveys are not known. A total of 157 normotensive subjects aged 18 to 35 years, selected at random in a cross-sectional population-based survey, answered a structured questionnaire. They were classified as strongly predisposed to hypertension when two or more first-degree relatives had a diagnosis of hypertension. Anthropometric parameters were obtained and sitting blood pressure was determined with aneroid sphygmomanometers. Sodium and potassium excretion was measured by flame spectrophotometry in an overnight urine sample. A positive correlation between blood pressure and urinary sodium excretion was detected only in the group of individuals strongly predisposed to hypertension, both for systolic blood pressure (r = 0.51, P<0.01) and diastolic blood pressure (r = 0.50, P<0.01). In a covariance analysis, after controlling for age, skin color and body mass index, individuals strongly predisposed to hypertension who excreted amounts of sodium above the median of the entire sample had higher systolic and diastolic blood pressure than subjects classified into the remaining conditions. The influence of familial predisposition to hypertension on the association between salt intake and blood pressure may be an additional explanation for the weak association between urinary sodium excretion and blood pressure observed in within-population studies, since it can influence the association between salt consumption and blood pressure in some but not all inhabitants.

For whom the cruiser rolls: a retrospective analysis of factors relating to recidivism in a sample of graduates from a court-ordered alcohol education and awareness program

This study investigated, retrospectively, whether recidivism in a sample of court-ordered'graduates of an alcohol education and awareness program could be predicted. This alcohol education program was based on adult education principles and was philosophically akin to the thoughts of Drs. Jack Mezirow, Stephen Brookfield, and Patricia Cranton. Data on the sample of 214 Halton IDEA (Impaired Driver Education and Awareness) graduates were entered into a spread sheet. Descriptive statistics were generated. Each of the 214 program graduates had taken several tests during the course of the IDEA program. These tests measured knowledge, attitude about impaired driving, and degree of alcohol involvement. Test scores were analyzed to determine whether those IDEA graduates who recidivated differed in any measurable way from those who had no further criminal convictions after a period of at least three years. Their criminal records were obtained from the Canadian Police Information Centre (CPIC). Those program graduates who reoffended were compared to the vast majority who did not reoffend. Results of the study indicated that there was no way to determine who would recidivate from the data that were collected. Further studies could use a qualitative model. Follow-up interviews could be used to determine what impact, if any, attendance at the IDEA program had on the life of the graduates.

Finite-Sample Diagnostics for Multivariate Regressions with Applications to Linear Asset Pricing Models

In this paper, we propose several finite-sample specification tests for multivariate linear regressions (MLR) with applications to asset pricing models. We focus on departures from the assumption of i.i.d. errors assumption, at univariate and multivariate levels, with Gaussian and non-Gaussian (including Student t) errors. The univariate tests studied extend existing exact procedures by allowing for unspecified parameters in the error distributions (e.g., the degrees of freedom in the case of the Student t distribution). The multivariate tests are based on properly standardized multivariate residuals to ensure invariance to MLR coefficients and error covariances. We consider tests for serial correlation, tests for multivariate GARCH and sign-type tests against general dependencies and asymmetries. The procedures proposed provide exact versions of those applied in Shanken (1990) which consist in combining univariate specification tests. Specifically, we combine tests across equations using the MC test procedure to avoid Bonferroni-type bounds. Since non-Gaussian based tests are not pivotal, we apply the “maximized MC” (MMC) test method [Dufour (2002)], where the MC p-value for the tested hypothesis (which depends on nuisance parameters) is maximized (with respect to these nuisance parameters) to control the test’s significance level. The tests proposed are applied to an asset pricing model with observable risk-free rates, using monthly returns on New York Stock Exchange (NYSE) portfolios over five-year subperiods from 1926-1995. Our empirical results reveal the following. Whereas univariate exact tests indicate significant serial correlation, asymmetries and GARCH in some equations, such effects are much less prevalent once error cross-equation covariances are accounted for. In addition, significant departures from the i.i.d. hypothesis are less evident once we allow for non-Gaussian errors.

Monte Carlo Tests with Nuisance Parameters: A General Approach to Finite-Sample Inference and Nonstandard Asymptotics

The technique of Monte Carlo (MC) tests [Dwass (1957), Barnard (1963)] provides an attractive method of building exact tests from statistics whose finite sample distribution is intractable but can be simulated (provided it does not involve nuisance parameters). We extend this method in two ways: first, by allowing for MC tests based on exchangeable possibly discrete test statistics; second, by generalizing the method to statistics whose null distributions involve nuisance parameters (maximized MC tests, MMC). Simplified asymptotically justified versions of the MMC method are also proposed and it is shown that they provide a simple way of improving standard asymptotics and dealing with nonstandard asymptotics (e.g., unit root asymptotics). Parametric bootstrap tests may be interpreted as a simplified version of the MMC method (without the general validity properties of the latter).