Contributions to linear discriminant analysis with applications to growth curves

Contributions to linear discriminant analysis with applications to growth curves

Author: Edward Kanuti Ngailo

Publisher: Linköping University Electronic Press

Published: 2020-05-06

Total Pages: 47

ISBN-13: 9179298567

DOWNLOAD EBOOK

This thesis concerns contributions to linear discriminant analysis with applications to growth curves. Firstly, we present the linear discriminant function coefficients in a stochastic representation using random variables from the standard univariate distributions. We apply the characterized distribution in the classification function to approximate the classification error rate. The results are then extended to large dimension asymptotics under assumption that the dimension p of the parameter space increases together with the sample size n to infinity such that the ratio converges to a positive constant c (0, 1). Secondly, the thesis treats repeated measures data which correspond to multiple measurements that are taken on the same subject at different time points. We develop a linear classification function to classify an individual into one out of two populations on the basis of the repeated measures data that when the means follow a growth curve structure. The growth curve structure we first consider assumes that all treatments (groups) follows the same growth profile. However, this is not necessarily true in general and the problem is extended to linear classification where the means follow an extended growth curve structure, i.e., the treatments under the experimental design follow different growth profiles. At last, a function of the inverse Wishart matrix and a normal distribution finds its application in portfolio theory where the vector of optimal portfolio weights is proportional to the product of the inverse sample covariance matrix and a sample mean vector. Analytical expressions for higher order moments and non-central moments of the portfolio weights are derived when the returns are assumed to be independently multivariate normally distributed. Moreover, the expressions for the mean, variance, skewness and kurtosis of specific estimated weights are obtained. The results are complemented using a Monte Carlo simulation study, where data from the multivariate normal and t-distributions are discussed. Den här avhandlingen studerar diskriminantanalys, klassificering av tillväxtkurvor och portföljteori. Diskriminantanalys och klassificering är flerdimensionella tekniker som används för att separera olika mängder av objekt och för att tilldela nya objekt till redan definierade grupper (så kallade klasser). En klassisk metod är att använda Fishers linjära diskriminantfunktion och när alla parametrar är kända så kan man enkelt beräkna sannolikheterna för felklassificering. Tyvärr är så sällan fallet, utan parametrarna måste skattas från data, och då blir Fishers linjära diskriminantfunktion en funktion av en Wishartmatris och multivariat normalfördelade vektorer. I den här avhandlingen studerar vi hur man kan approximativt beräkna sannolikheten för felklassificering under antagande att dimensionen på parameterrummet ökar tillsammans med antalet observationer genom att använda en särskild stokastisk representation av diskriminantfunktionen. Upprepade mätningar över tiden på samma individ eller objekt går att modellera med så kallade tillväxtkurvor. Vid klassificering av tillväxtkurvor, eller rättare sagt av upprepade mätningar för en ny individ, bör man ta tillvara på både den spatiala- och temporala informationen som finns hos dessa observationer. Vi vidareutvecklar Fishers linjära diskriminantfunktion att passa för upprepade mätningar och beräknar asymptotiska sannolikheter för felklassificering. Till sist kan man notera att snarlika funktioner av Wishartmatriser och multivariat normalfördelade vektorer dyker upp när man vill beräkna de optimala vikterna i portföljteori. Genom en stokastisk representation studerar vi egenskaperna hos portföljvikterna och gör dessutom en simuleringsstudie för att förstå vad som händer när antagandet om normalfördelning inte är uppfyllt.


Discriminant Analysis and Applications

Discriminant Analysis and Applications

Author: T. Cacoullos

Publisher: Academic Press

Published: 2014-05-10

Total Pages: 455

ISBN-13: 1483268713

DOWNLOAD EBOOK

Discriminant Analysis and Applications comprises the proceedings of the NATO Advanced Study Institute on Discriminant Analysis and Applications held in Kifissia, Athens, Greece in June 1972. The book presents the theory and applications of Discriminant analysis, one of the most important areas of multivariate statistical analysis. This volume contains chapters that cover the historical development of discriminant analysis methods; logistic and quasi-linear discrimination; and distance functions. Medical and biological applications, and computer graphical analysis and graphical techniques for multidimensional data are likewise discussed. Statisticians, mathematicians, and biomathematicians will find the book very interesting.


Contributions to Discriminant Analysis of Cross-sectional and Longitudinal Data with Applications

Contributions to Discriminant Analysis of Cross-sectional and Longitudinal Data with Applications

Author: Alice M. Hinton

Publisher:

Published: 2014

Total Pages:

ISBN-13:

DOWNLOAD EBOOK

There are a variety of methods available to classify an object into one of two populations. Here, the method of discriminant analysis is considered in the cross-sectional and the longitudinal setting with a structured multivariate normal model. The generalized likelihood ratio change detection algorithm is also investigated as an alternative to methods based on discriminant analysis in the longitudinal setting. Traditionally, discriminant functions are developed to classify a new observation from a cross-sectional dataset into a population. An error is made when the observation is incorrectly classified. In the literature, several parametric and empirical methods of estimating these misclassification probabilities have been proposed. The performance of six parametric and three empirical misclassification probability estimators are compared. It is found that the parametric methods, which rely on an assumption of normality, generally outperform the empirical methods when a linear discriminant function is used for classification and the data originate from normal populations. The preferred parametric method depends on the size of the training dataset and the parameters of the populations, particularly the distance between the means. The empirical methods are preferred only when the two populations are well separated and the variances are significantly different.


Discriminant Analysis

Discriminant Analysis

Author: William R. Klecka

Publisher: SAGE Publications, Incorporated

Published: 1980-08-01

Total Pages: 78

ISBN-13: 9780803914919

DOWNLOAD EBOOK

These procedures, collectively known as discriminant analysis, allow a researcher to study the difference between two or more groups of objects with respect to several variables simultaneously, determining whether meaningful differences exist between the groups and identifying the discriminating power of each variable.


Discriminant Analysis

Discriminant Analysis

Author: Peter A. Lachenbruch

Publisher:

Published: 1975

Total Pages: 146

ISBN-13:

DOWNLOAD EBOOK

Basic ideas of discriminant analysis; Evaluating a discriminant function; Robustness of the linear discriminant function; Nonnormal and nonparametric methods; Multiple-group problems; Miscellaneous problems.


Discrete Discriminant Analysis

Discrete Discriminant Analysis

Author: Matthew Goldstein

Publisher: John Wiley & Sons

Published: 1978

Total Pages: 206

ISBN-13:

DOWNLOAD EBOOK

The linear discriminant function; Discrete classification models; Error rates and the problem of bias; The variable-selection problem; Special topics; Computer programs.


Applied Univariate, Bivariate, and Multivariate Statistics

Applied Univariate, Bivariate, and Multivariate Statistics

Author: Daniel J. Denis

Publisher: John Wiley & Sons

Published: 2021-04-01

Total Pages: 578

ISBN-13: 1119583012

DOWNLOAD EBOOK

AN UPDATED GUIDE TO STATISTICAL MODELING TECHNIQUES USED IN THE SOCIAL AND NATURAL SCIENCES This revised and updated second edition of Applied Univariate, Bivariate, and Multivariate Statistics: Understanding Statistics for Social and Natural Scientists, with Applications in SPSS and R contains an accessible introduction to statistical modeling techniques commonly used in the social and natural sciences. The text offers a blend of statistical theory and methodology and reviews both the technical and theoretical aspects of good data analysis. Featuring applied resources at various levels, the book includes statistical techniques using software packages such as R and SPSS®. To promote a more in-depth interpretation of statistical techniques across the sciences, the book surveys some of the technical arguments underlying formulas and equations. The second edition has been designed to be more approachable by minimizing theoretical or technical jargon and maximizing conceptual understanding with easy-to-apply software examples. This important text: Offers demonstrations of statistical techniques using software packages such as R and SPSS® Contains examples of hypothetical and real data with statistical analyses Provides historical and philosophical insights into many of the techniques used in modern science Includes a companion website that features further instructional details, additional data sets, and solutions to selected exercises Written for students of social and applied sciences, Applied Univariate, Bivariate, and Multivariate Statistics, Second Edition offers a thorough introduction to the world of statistical modeling techniques in the sciences.


Stock Identification Methods

Stock Identification Methods

Author: Lisa A. Kerr

Publisher: Elsevier

Published: 2004-10-15

Total Pages: 735

ISBN-13: 0080470432

DOWNLOAD EBOOK

Stock Identification Methods provides a comprehensive review of the various disciplines used to study the population structure of fishery resources. It represents the worldwide experience and perspectives of experts on each method, assembled through a working group of the International Council for the Exploration of the Sea. The book is organized to foster interdisciplinary analyses and conclusions about stock structure, a crucial topic for fishery science and management. Technological advances have promoted the development of stock identification methods in many directions, resulting in a confusing variety of approaches. Based on central tenets of population biology and management needs, Stock Identification Methods offers a unified framework for understanding stock structure by promoting an understanding of the relative merits and sensitivities of each approach. * Describes eighteen distinct approaches to stock identification grouped into sections on life history traits, environmental signals, genetic analyses, and applied marks* Features experts' reviews of benchmark case studies, general protocols, and the strengths and weaknesses of each identification method* Reviews statistical techniques for exploring stock patterns, testing for differences among putative stocks, stock discrimination, and stock composition analysis* Focuses on the challenges of interpreting data and managing mixed-stock fisheries