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.

Exact Skewness-Kurtosis Tests for Multivariate Normality and Goodness-of-fit in Multivariate Regressions with Application to Asset Pricing Models

We study the problem of testing the error distribution in a multivariate linear regression (MLR) model. The tests are functions of appropriately standardized multivariate least squares residuals whose distribution is invariant to the unknown cross-equation error covariance matrix. Empirical multivariate skewness and kurtosis criteria are then compared to simulation-based estimate of their expected value under the hypothesized distribution. Special cases considered include testing multivariate normal, Student t; normal mixtures and stable error models. In the Gaussian case, finite-sample versions of the standard multivariate skewness and kurtosis tests are derived. To do this, we exploit simple, double and multi-stage Monte Carlo test methods. For non-Gaussian distribution families involving nuisance parameters, confidence sets are derived for the the nuisance parameters and the error distribution. The procedures considered are evaluated in a small simulation experi-ment. Finally, the tests 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.

Exact Multivariate Tests of Asset Pricing Models with Stable Asymmetric Distributions

In this paper, we propose exact inference procedures for asset pricing models that can be formulated in the framework of a multivariate linear regression (CAPM), allowing for stable error distributions. The normality assumption on the distribution of stock returns is usually rejected in empirical studies, due to excess kurtosis and asymmetry. To model such data, we propose a comprehensive statistical approach which allows for alternative - possibly asymmetric - heavy tailed distributions without the use of large-sample approximations. The methods suggested are based on Monte Carlo test techniques. Goodness-of-fit tests are formally incorporated to ensure that the error distributions considered are empirically sustainable, from which exact confidence sets for the unknown tail area and asymmetry parameters of the stable error distribution are derived. Tests for the efficiency of the market portfolio (zero intercepts) which explicitly allow for the presence of (unknown) nuisance parameter in the stable error distribution are derived. The methods proposed are applied to monthly returns on 12 portfolios of the New York Stock Exchange over the period 1926-1995 (5 year subperiods). We find that stable possibly skewed distributions provide statistically significant improvement in goodness-of-fit and lead to fewer rejections of the efficiency hypothesis.

Le développement économique régional, la politique et les disparités régionales des grands territoires : le cas du Canada et de la Chine

Résumé Depuis le début des années 1990, la recherche sur le développement régional a pris une importance considérable dans les disciplines de l’économie et de la géographie dans la plupart des pays. De nombreuses études ont été consacrées à ce sujet et l’on constate une approche analytique de plus en plus sophistiquée. Que les économies pauvres ont tendance à converger vers les pays riches, ou bien à diverger au fil du temps est une question qui a attiré l'attention des décideurs et des universitaires depuis quelques décennies. Convergence ou de divergence économique est un sujet d'intérêt et de débat, non seulement pour valider ou non les deux modèles principaux de croissance qui sont considérés comme concurrent (l’approche néo-classique et celle des approches de croissance endogène), mais aussi pour ses implications pour les publiques politiques. En se basant sur une analyse des politiques de développement régional et des analyses statistiques de la convergence et des disparités régionales, les objectifs de cette thèse sont de tenter de fournir une explication des différents processus et des modèles de développement économique régional poursuivis dans le cas de territoires immenses en utilisant le Canada et la Chine comme études de cas, d'entreprendre une analyse des différents facteurs et des forces motrices qui sous-tendent le développement régional dans ces deux pays, et d'explorer à la fois les réussites et les échecs apparents dans les politiques de développement régional en comparant et contrastant les expériences de développement régional et les modèles de ces deux pays. A fin d'atteindre cet objectif, la recherche utilise une approche multi-scalaire et des méthodes de mesure multidimensionnelle dans le cadre des analyses sur les disparités « régionales » entre les macro régions (sous-ensembles de provinces) des deux pays, des provinces et des régions urbaines sélectionnées, dans le but ultime d’identifier des problèmes existants en termes de développement régional et de pouvoir proposer des solutions. Les étapes principales de la recherche sont : 1. La cueillette des données statistiques pour le Canada et la Chine (incluant les provinces de Québec et de Xinjiang) pour une gamme d’indicateurs (voir ci-dessous). 2. D’entreprendre une analyse de chaque dimension dans les deux juridictions: Population (p.ex. composition, structure, changement); Ressources (p. ex. utilisation, exploitation de l’énergie); Environnement (p.ex. la pollution); et le Développement socioéconomique (p.ex. le développement et la transformation des secteurs clé, et les modèles de développement rural et urbain), et les disparités changeantes par rapport à ces dimensions. 3. La définition d’une typologie de différents types de région en fonction de leurs trajectoires de développement, ce qui servira pour critiquer l’hypothèse centre-périphérie. 4. Le choix d’une région métropolitaine dans chaque juridiction (province). 5. D’entreprendre une analyse temporelle des événements clé (politiques, investissements) dans chaque région et les facteurs impliqués dans chaque événement, en utilisant l’information documentaire générale et des agences institutionnelles impliqués actuellement et dans un passée récent. Cette étude a tenté d'expliquer les schémas et les processus des deux économies, ainsi que la présentation d'études de cas qui illustrent et examinent les différences dans les deux économies à partir de l’échelle nationale jusqu’au niveau régional et provincial et aussi pour certaines zones urbaines. Cette étude a essayé de répondre aux questions de recherche comme: Est-il vrai que les pays avec des plus grandes territoires sont associés avec des plus grandes disparités interrégionales? Quel est le résultat des comparaisons entre pays développés et pays en développement? Quels sont les facteurs les plus importants dans le développement économique de vastes territoires dans les pays développés et pays en développement? Quel est le mécanisme de convergence et de divergence dans les pays développés et, respectivement, les pays en développement? Dans l'introduction à la thèse, le cadre général de l'étude est présenté, suivie dans le chapitre 1 d'une discussion sur les théories et les concepts utilisés dans la littérature théorique principale qui est pertinent à l'étude. Le chapitre 2 décrit la méthodologie de recherche. Le chapitre 3 présente une vue d'ensemble des politiques de développement économique régional et les programmes du Canada et de la Chine dans des périodes différentes à différentes échelles. Au chapitre 4, la convergence des deux pays à l'échelle nationale et la convergence provinciale pour chaque pays sont examinés en utilisant différentes méthodes de mesure telles que les méthodes traditionnelles, la convergence bêta et la convergence sigma. Dans le chapitre le plus complexe, le chapitre 5, les analyses comparatives sont présentées à l'aide de données statistiques, à partir des analyses des cas régionaux et provinciaux retenus des deux pays. Au chapitre 6, ces dispositions sont complétées par une analyse des régions urbaines choisies, qui permet également des aperçus sur les régions les plus périphériques. Dans la recherche proposée pour cette thèse, la politique, la population, le revenu, l'emploi, la composition industrielle, l'investissement, le commerce et le facteur de la migration sont également pris en compte comme facteurs importants de l'analyse régionale compte tenu de la superficie du territoire des deux pays et les différences de population entre eux. Cette thèse a évalué dans quelle mesure les politiques gouvernementales ont réussi à induire la convergence régionale ou ont encore ont creusé davantage les disparités régionales, ce qui implique nécessairement une évaluation de la durabilité des patrons et des programmes de développement régional. Cette étude a également mis l'accent sur les disparités régionales et la politique de développement régional, les comparaisons entre pays, pour mesurer la convergence entre les pays et entre les régions, y compris l'analyse spatiale, d'identifier les facteurs les plus actifs tels que la population, les ressources, la politique, l'urbanisation, les migrations, l'ouverture économique et leurs différents rôles dans le développement économique de ces grands territoires (au Canada et Chine). Les résultats empiriques et les processus de convergence et de divergence offrent un cadre intéressant pour l'examen de la trajectoire de développement régionales et les disparités régionales dans les deux économies. L'approche adoptée a révélé les différentes mosaïques complexes du développement régional dans les deux pays. Les résultats de cette étude ont démontré que la disparité en termes de revenu régional est une réalité dans chaque zone géographique, et que les causes sont nombreuses et complexes. Les deux économies ont certains parallèles dans la mise en œuvre des politiques de développement économique régional, mais il existe des différences importantes aussi et elles se sont développées à différentes vitesses. Les deux économies se sont développées depuis la Seconde Guerre mondiale, mais la Chine a connu une croissance rapide que le Canada comme témoignent de nombreux indicateurs depuis 1980. Cependant, la Chine est maintenant confrontée à un certain nombre de problèmes économiques et sociaux, y compris les disparités régionales marquées, un fossé toujours croissant entre les revenus ruraux et urbains, une population vieillissante, le chômage, la pauvreté et la dégradation rapide de l'environnement avec toujours plus de demandes en énergie. Le développement économique régional en Chine est plus déséquilibré. Le Canada accuse un degré de disparités régionales et provinciales moins important que la Chine. Dans les cas provinciaux, il existe d'importantes différences et de disparités dans la structure économique et spatiale du Québec et du Xinjiang. Les disparités infra provinciales sont plus grandes que celles à l’échelle des provinces et des grandes régions (des sous-ensembles de provinces). Les mécanismes de convergence et de divergence dans les deux pays sont différents. Les résultats empiriques et les processus de convergence et de divergence offrent un cadre intéressant pour l'examen de la trajectoire de développement régionale et les disparités régionales dans les deux économies. Cette étude démontre également que l'urbanisation (les métropoles et les villes) s’avère être le facteur le plus actif et contribue à l'économie régionale dans ces grands territoires. L'ouverture a joué un rôle important dans les économies des deux pays. La migration est un facteur majeur dans la stimulation de l'économie des deux pays mais de façons différentes. Les résultats empiriques démontrent que les disparités régionales ne peuvent pas être évitées et elles existent presque partout. Il n'y a pas une formule universelle et de politiques spécifiques sont de mise pour chaque région. Mais il semble possible pour les décideurs politiques nationaux et régionaux d’essayer de maintenir l'écart à une échelle raisonnable pour éviter l'instabilité sociale.

Statistical model based 3D shape prediction of postoperative trunks for non-invasive scoliosis surgery planning

One of the major concerns of scoliosis patients undergoing surgical treatment is the aesthetic aspect of the surgery outcome. It would be useful to predict the postoperative appearance of the patient trunk in the course of a surgery planning process in order to take into account the expectations of the patient. In this paper, we propose to use least squares support vector regression for the prediction of the postoperative trunk 3D shape after spine surgery for adolescent idiopathic scoliosis. Five dimensionality reduction techniques used in conjunction with the support vector machine are compared. The methods are evaluated in terms of their accuracy, based on the leave-one-out cross-validation performed on a database of 141 cases. The results indicate that the 3D shape predictions using a dimensionality reduction obtained by simultaneous decomposition of the predictors and response variables have the best accuracy.

Further studies on the process of coastal upwelling along the south west coast of India - possible correlations with biological productivity

An attempt is made to study the possible relationship between the process of upwelling and zooplankton biomass in the shelf weters along the south west coast of India between Cape comorin and Ratnagiri based on oceanographic and Zooplankton data collected by the erstwhile FAO/UNDP Pelagic Fishery Project,Cochin between 1973 and 1978. Different factors such as the depth from which the bottom waters are induced upwards during the process of upwelling,the depth to which the bottom waters are drawn, vertical velocity of upwelling and the resultant zooplankton productivity were considered while arriving at the deductions. Except for nutrients and phytoplankton productivity, for which simultaneous data is lacking, all the major factors were taken into consideration before cocluding- xon positive/negative correlation.

Bivariate semi-Pareto distributions and processes

A bivariate semi-Pareto distribution is introduced and characterized using geometric minimization. Autoregressive minification models for bivariate random vectors with bivariate semi-Pareto and bivariate Pareto distributions are also discussed. Multivariate generalizations of the distributions and the processes are briefly indicated.

Process Optimization for Mass Production of Marine Yeast Candida sp. S 27 and its Nutritional Characterization

The main source of protein for human and animal consumption is from the agricultural sector, where the production is vulnerable to diseases, fluctuations in climatic conditions and deteriorating hydrological conditions due to water pollution. Therefore Single Cell Protein (SCP) production has evolved as an excellent alternative. Among all sources of microbial protein, yeast has attained global acceptability and has been preferred for SCP production. The screening and evaluation of nutritional and other culture variables of microorganisms are very important in the development of a bioprocess for SCP production. The application of statistical experimental design in bioprocess development can result in improved product yields, reduced process variability, closer confirmation of the output response to target requirements and reduced development time and overall cost.The present work was undertaken to develop a bioprocess technology for the mass production of a marine yeast, Candida sp.S27. Yeasts isolated from the offshore waters of the South west coast of India and maintained in the Microbiology Laboratory were subjected to various tests for the selection of a potent strain for biomass production. The selected marine yeast was identified based on ITS sequencing. Biochemical/nutritional characterization of Candida sp.S27 was carried out. Using Response Surface Methodology (RSM) the process parameters (pH, temperature and salinity) were optimized. For mass production of yeast biomass, a chemically defined medium (Barnett and Ingram, 1955) and a crude medium (Molasses-Yeast extract) were optimized using RSM. Scale up of biomass production was done in a Bench top Fermenter using these two optimized media. Comparative efficacy of the defined and crude media were estimated besides nutritional evaluation of the biomass developed using these two optimized media.

Statistical signal extraction using stable processes

The standard models for statistical signal extraction assume that the signal and noise are generated by linear Gaussian processes. The optimum filter weights for those models are derived using the method of minimum mean square error. In the present work we study the properties of signal extraction models under the assumption that signal/noise are generated by symmetric stable processes. The optimum filter is obtained by the method of minimum dispersion. The performance of the new filter is compared with their Gaussian counterparts by simulation.

Applying Data mining using Statistical Techniques for Career Selection

For years, choosing the right career by monitoring the trends and scope for different career paths have been a requirement for all youngsters all over the world. In this paper we provide a scientific, data mining based method for job absorption rate prediction and predicting the waiting time needed for 100% placement, for different engineering courses in India. This will help the students in India in a great deal in deciding the right discipline for them for a bright future. Information about passed out students are obtained from the NTMIS ( National technical manpower information system ) NODAL center in Kochi, India residing in Cochin University of science and technology

English to Malayalam Translation: A Statistical Approach

This paper underlines a methodology for translating text from English into the Dravidian language, Malayalam using statistical models. By using a monolingual Malayalam corpus and a bilingual English/Malayalam corpus in the training phase, the machine automatically generates Malayalam translations of English sentences. This paper also discusses a technique to improve the alignment model by incorporating the parts of speech information into the bilingual corpus. Removing the insignificant alignments from the sentence pairs by this approach has ensured better training results. Pre-processing techniques like suffix separation from the Malayalam corpus and stop word elimination from the bilingual corpus also proved to be effective in training. Various handcrafted rules designed for the suffix separation process which can be used as a guideline in implementing suffix separation in Malayalam language are also presented in this paper. The structural difference between the English Malayalam pair is resolved in the decoder by applying the order conversion rules. Experiments conducted on a sample corpus have generated reasonably good Malayalam translations and the results are verified with F measure, BLEU and WER evaluation metrics

A framework for translating English text into Malayalam using statistical models

A methodology for translating text from English into the Dravidian language, Malayalam using statistical models is discussed in this paper. The translator utilizes a monolingual Malayalam corpus and a bilingual English/Malayalam corpus in the training phase and generates automatically the Malayalam translation of an unseen English sentence. Various techniques to improve the alignment model by incorporating the morphological inputs into the bilingual corpus are discussed. Removing the insignificant alignments from the sentence pairs by this approach has ensured better training results. Pre-processing techniques like suffix separation from the Malayalam corpus and stop word elimination from the bilingual corpus also proved to be effective in producing better alignments. Difficulties in translation process that arise due to the structural difference between the English Malayalam pair is resolved in the decoding phase by applying the order conversion rules. The handcrafted rules designed for the suffix separation process which can be used as a guideline in implementing suffix separation in Malayalam language are also presented in this paper. Experiments conducted on a sample corpus have generated reasonably good Malayalam translations and the results are verified with F measure, BLEU and WER evaluation metrics

Handling OOV Words in Phrase-Based Statistical Machine Translation for Malayalam

Statistical Machine Translation (SMT) is one of the potential applications in the field of Natural Language Processing. The translation process in SMT is carried out by acquiring translation rules automatically from the parallel corpora. However, for many language pairs (e.g. Malayalam- English), they are available only in very limited quantities. Therefore, for these language pairs a huge portion of phrases encountered at run-time will be unknown. This paper focuses on methods for handling such out-of-vocabulary (OOV) words in Malayalam that cannot be translated to English using conventional phrase-based statistical machine translation systems. The OOV words in the source sentence are pre-processed to obtain the root word and its suffix. Different inflected forms of the OOV root are generated and a match is looked up for the word variants in the phrase translation table of the translation model. A Vocabulary filter is used to choose the best among the translations of these word variants by finding the unigram count. A match for the OOV suffix is also looked up in the phrase entries and the target translations are filtered out. Structuring of the filtered phrases is done and SMT translation model is extended by adding OOV with its new phrase translations. By the results of the manual evaluation done it is observed that amount of OOV words in the input has been reduced considerably

Studies on nitrifying microorganisms in Cochin Estuary and adjacent coastal waters

This thesis entitled “Studies on Nitrifying Microorganisms in Cochin Estuary and Adjacent Coastal Waters” reports for the first time the spatial andtemporal variations in the abundance and activity of nitrifiers (Ammonia oxidizingbacteria-AOB; Nitrite oxidizing bacteria- NOB and Ammonia oxidizing archaea-AOA) from the Cochin Estuary (CE), a monsoon driven, nutrient rich tropicalestuary along the southwest coast of India. To fulfil the above objectives, field observations were carried out for aperiod of one year (2011) in the CE. Surface (1 m below surface) and near-bottomwater samples were collected from four locations (stations 1 to 3 in estuary and 4 in coastal region), covering pre-monsoon, monsoon and post-monsoon seasons. Station 1 is a low saline station (salinity range 0-10) with high freshwater influx While stations 2 and 3 are intermediately saline stations (salinity ranges 10-25). Station 4 is located ~20 km away from station 3 with least influence of fresh water and is considered as high saline (salinity range 25- 35) station. Ambient physicochemical parameters like temperature, pH, salinity, dissolved oxygen (DO), Ammonium, nitrite, nitrate, phosphate and silicate of surface and bottom waters were measured using standard techniques. Abundance of Eubacteria, total Archaea and ammonia and nitrite oxidizing bacteria (AOB and NOB) were quantified using Fluorescent in situ Hybridization (FISH) with oligonucleotide probes labeled withCy3. Community structure of AOB and AOA was studied using PCR Denaturing Gradient Gel Electrophoresis (DGGE) technique. PCR products were cloned and sequenced to determine approximate phylogenetic affiliations. Nitrification rate in the water samples were analyzed using chemical NaClO3 (inhibitor of nitrite oxidation), and ATU (inhibitor of ammonium oxidation). Contribution of AOA and AOB in ammonia oxidation process was measured based on the recovered ammonia oxidation rate. The contribution of AOB and AOA were analyzed after inhibiting the activities of AOB and AOA separately using specific protein inhibitors. To understand the factors influencing or controlling nitrification, various statistical tools were used viz. Karl Pearson’s correlation (to find out the relationship between environmental parameters, bacterial abundance and activity), three-way ANOVA (to find out the significant variation between observations), Canonical Discriminant Analysis (CDA) (for the discrimination of stations based on observations), Multivariate statistics, Principal components analysis (PCA) and Step up multiple regression model (SMRM) (First order interaction effects were applied to determine the significantly contributing biological and environmental parameters to the numerical abundance of nitrifiers). In the CE, nitrification is modulated by the complex interplay between different nitrifiers and environmental variables which in turn is dictated by various hydrodynamic characteristics like fresh water discharge and seawater influx brought in by river water discharge and flushing. AOB in the CE are more adapted to varying environmental conditions compared to AOA though the diversity of AOA is higher than AOB. The abundance and seasonality of AOB and NOB is influenced by the concentration of ammonia in the water column. AOB are the major players in modulating ammonia oxidation process in the water column of CE. The distribution pattern and seasonality of AOB and NOB in the CE suggest that these organisms coexist, and are responsible for modulating the entire nitrification process in the estuary. This process is fuelled by the cross feeding among different nitrifiers, which in turn is dictated by nutrient levels especially ammonia. Though nitrification modulates the increasing anthropogenic ammonia concentration the anthropogenic inputs have to be controlled to prevent eutrophication and associated environmental changes.