Finite-Sample Inference Methods for Simultaneous Equations and Models with Unobserved and Generated Regressors

We propose finite sample tests and confidence sets for models with unobserved and generated regressors as well as various models estimated by instrumental variables methods. The validity of the procedures is unaffected by the presence of identification problems or \"weak instruments\", so no detection of such problems is required. We study two distinct approaches for various models considered by Pagan (1984). The first one is an instrument substitution method which generalizes an approach proposed by Anderson and Rubin (1949) and Fuller (1987) for different (although related) problems, while the second one is based on splitting the sample. The instrument substitution method uses the instruments directly, instead of generated regressors, in order to test hypotheses about the \"structural parameters\" of interest and build confidence sets. The second approach relies on \"generated regressors\", which allows a gain in degrees of freedom, and a sample split technique. For inference about general possibly nonlinear transformations of model parameters, projection techniques are proposed. A distributional theory is obtained under the assumptions of Gaussian errors and strictly exogenous regressors. We show that the various tests and confidence sets proposed are (locally) \"asymptotically valid\" under much weaker assumptions. The properties of the tests proposed are examined in simulation experiments. In general, they outperform the usual asymptotic inference methods in terms of both reliability and power. Finally, the techniques suggested are applied to a model of Tobin’s q and to a model of academic performance.

Simulation-Based Finite-and Large-sample Inference Methods in Multivariate Regressions and Seemingly Unrelated Regressions

In the context of multivariate regression (MLR) and seemingly unrelated regressions (SURE) models, it is well known that commonly employed asymptotic test criteria are seriously biased towards overrejection. in this paper, we propose finite-and large-sample likelihood-based test procedures for possibly non-linear hypotheses on the coefficients of MLR and SURE systems.

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).

Finite-Sample Simulation-Based Inference in VAR Models with Applications to Order Selection and Causality Testing

Statistical tests in vector autoregressive (VAR) models are typically based on large-sample approximations, involving the use of asymptotic distributions or bootstrap techniques. After documenting that such methods can be very misleading even with fairly large samples, especially when the number of lags or the number of equations is not small, we propose a general simulation-based technique that allows one to control completely the level of tests in parametric VAR models. In particular, we show that maximized Monte Carlo tests [Dufour (2002)] can provide provably exact tests for such models, whether they are stationary or integrated. Applications to order selection and causality testing are considered as special cases. The technique developed is applied to quarterly and monthly VAR models of the U.S. economy, comprising income, money, interest rates and prices, over the period 1965-1996.

How do social interactions with a significant other affect PTSD symptoms? An empirical investigation with a clinical sample.

Social support and coping are both related to posttraumatic stress disorder (PTSD) symptoms, but the mechanisms underlying their relationships remain unclear. This study explores these relationships by examining the perceived frequency of supportive and countersupportive interactions with a significant other in PTSD patients. Ninety-six participants with PTSD were recruited and completed questionnaires assessing social interactions, ways of coping, and PTSD symptoms. Associations of social interactions (r2 = 4.1%–7.9%, p < .05) and coping (r2 = 15.9%– 16.5%, p < .001) with symptoms were independent, and suggested a direct association between social interactions and PTSD. Countersupportive interactions were more associated to symptoms than supportive interactions. Our findings suggest the development of psychotherapies that integrate social support interventions.

Mental health treatment seeking by military members with posttraumatic stress disorder : findings on rates, characteristics, and predictors from a nationally representative Canadian military sample

Objective: The goal of this study was to identify rates, characteristies, and predictors of mental health treatment seeking by military members with posttraumatic stress disorder (PTSD). Method: Our sample was drawn from the 2002 Canadian Community Health Survey-Canadian Forces Supplement (CCHS-CF) dataset. The CCHS-CF is the first epidemiologic survey of PTSD and other mental health conditions in the Canadian military and includes 8441 nationally representative Canadian Forces (CF) members. Of those, 549 who met the criteria for lifetime PTSD were included in our analyses. To identify treatment rates and characteristics, we examined frequency of treatment contact by professional and facility type. To identify predictors of treatment seeking, we conducted a binary logistic regression with lifetime treatment seeking as the outcome variable. Results: About two-thirds of those with PTSD consulted with a professional regarding mental health problems. The most frequently consulted professionals, during both the last year and lifetime, included social workers and counsellors, medical doctors and general practitioners, and psychiatrists. Consultations during the last year most often took place in a CF facility. Treatment seeking was predicted by cumulative lifetime trauma exposure, index traumatic event type, PTSD symptom interference, and comorbid major depressive disorder. Those with comorbid depression were 3.75 times more likely to have sought treatment than those without. Conclusions: Although a significant portion of military members with PTSD sought mental health treatment, 1 in 3 never did. Trauma-related and illness and (or) need factors predicted treatment seeking. Of all the predictors of treatment seeking, comorbid depression most increased the likelihood of seeking treatment.

Échantillonnage des distributions continues non uniformes en précision arbitraire et protocole pour l'échantillonnage exact distribué des distributions discrètes quantiques

La thèse est divisée principalement en deux parties. La première partie regroupe les chapitres 2 et 3. La deuxième partie regroupe les chapitres 4 et 5. La première partie concerne l'échantillonnage de distributions continues non uniformes garantissant un niveau fixe de précision. Knuth et Yao démontrèrent en 1976 comment échantillonner exactement n'importe quelle distribution discrète en n'ayant recours qu'à une source de bits non biaisés indépendants et identiquement distribués. La première partie de cette thèse généralise en quelque sorte la théorie de Knuth et Yao aux distributions continues non uniformes, une fois la précision fixée. Une borne inférieure ainsi que des bornes supérieures pour des algorithmes génériques comme l'inversion et la discrétisation figurent parmi les résultats de cette première partie. De plus, une nouvelle preuve simple du résultat principal de l'article original de Knuth et Yao figure parmi les résultats de cette thèse. La deuxième partie concerne la résolution d'un problème en théorie de la complexité de la communication, un problème qui naquit avec l'avènement de l'informatique quantique. Étant donné une distribution discrète paramétrée par un vecteur réel de dimension N et un réseau de N ordinateurs ayant accès à une source de bits non biaisés indépendants et identiquement distribués où chaque ordinateur possède un et un seul des N paramètres, un protocole distribué est établi afin d'échantillonner exactement ladite distribution.

Studies on Fuzzy Graphs

In this thesis an attempt to develop the properties of basic concepts in fuzzy graphs such as fuzzy bridges, fuzzy cutnodes, fuzzy trees and blocks in fuzzy graphs have been made. The notion of complement of a fuzzy graph is modified and some of its properties are studied. Since the notion of complement has just been initiated, several properties of G and G available for crisp graphs can be studied for fuzzy graphs also. Mainly focused on fuzzy trees defined by Rosenfeld in [10] , several other types of fuzzy trees are defined depending on the acyclicity level of a fuzzy graph. It is observed that there are selfcentered fuzzy trees. Some operations on fuzzy graphs and prove that complement of the union two fuzzy graphs is the join of their complements and complement of the join of two fuzzy graphs is union of their complements. The study of fuzzy graphs made in this thesis is far from being complete. The wide ranging applications of graph theory and the interdisciplinary nature of fuzzy set theory, if properly blended together could pave a way for a substantial growth of fuzzy graph theory.

Fuzzy Absolutes and Related Topics

The study on the fuzzy absolutes and related topics. The different kinds of extensions especially compactification formed a major area of study in topology. Perfect continuous mappings always preserve certain topological properties. The concept of Fuzzy sets introduced by the American Cyberneticist L. A Zadeh started a revolution in every branch of knowledge and in particular in every branch of mathematics. Fuzziness is a kind of uncertainty and uncertainty of a symbol lies in the lack of well-defined boundaries of the set of objects to which this symbol belongs. Introduce an s-continuous mapping from a topological space to a fuzzy topological space and prove that the image of an H-closed space under an s-continuous mapping is f-H closed. Here also proved that the arbitrary product fi and sum of  fi of the s-continuous maps fi are also s-continuous. The original motivation behind the study of absolutes was the problem of characterizing the projective objects in the category of compact spaces and continuous functions.