...Multivariate Discriminant Analysis Priyanshi Gupta An Overview MDA is a statistical technique used to classify an observation into one of the several a priori groupings dependent on the observation’s individual characteristics. It is used primarily to classify and/or make predictions in the problems where dependent variable comes in qualitative form, for example, male or female, bankrupt or non-bankrupt etc. So the first step is to establish explicit group classifications. We have got observations coming from k groups. We are trying to look at what is the best way or best function in order to discriminate observations coming from different groups. Once such function is in place, we go to classification which basically is the problem of classification of a new observation into appropriate population using the discriminant function. So typically in such problems, once you have a set of data (called LEARNING set of data) with observations possibly coming from different populations are pre-classified, having predefined memberships to the groups. And based on the particular previously classified data, we create a discriminant function and can use it after proper calibration to classify a new observation to be coming from one of the groups. Discriminant analysis is used when groups are known a priori. Types of DA Problems 2 Group Problems... …regression can be used k-Group Problem (where k>=2)... …regression cannot be used if k>2 Example of a...
Words: 6778 - Pages: 28
...an email Cluster Analysis & Factor Analysis 325-711 Research Methods 2007 Lecturer: Jeromy Anglim “Of particular concern is the fairly routine use of a variation of exploratory factor analysis wherein the researcher uses principal components analysis (PCA), retains components with eigenvalues greater than 1 and uses varimax rotation, a bundle of procedures affectionately termed “Little Jiffy” …” Preacher, K. J., MacCallum, R. C. (2003). Repairing Tom Swift's Electric Factor Analysis Machine. Understanding Statistics, 2(1), 13-43. DECRIPTION: This session will first introduce students to factor analysis techniques including common factor analysis and principal components analysis. A factor analysis is a data reduction technique to summarize a number of original variables into a smaller set of composite dimensions, or factors. It is an important step in scale development and can be used to demonstrate construct validity of scale items. We will then move onto cluster analysis techniques. Cluster analysis groups individuals or objects into clusters so that objects in the same cluster are homogeneous and there is heterogeneity across clusters. This technique is often used to segment the data into similar, natural, groupings. For both analytical techniques, a focus will be on when to use the analytical technique, making reasoned decisions about options within each technique, and how to interpret the SPSS output. Slide 2 Overview • Factor Analysis & Principal Components...
Words: 10031 - Pages: 41
...Sweet Poison- Epidemic of the 21st Century! TEAM RISKY BUSINESS DEEPAK JOSHY JOBIN RAJ NEERAJA S RAVI TEJA POLINENI VIGNESH SANKAR J CONTENTS 1.0 INTRODUCTION ................................................................................................................................1 2.0 OBJECTIVES .......................................................................................................................................2 2.1 Decision Maker’s Problem: ....................................................................................................2 2.2 Market Research Problem: ....................................................................................................2 3.0 DESK RESEARCH .............................................................................................................................3 3.1 Purpose.....................................................................................................................................3 3.2 Procedure ................................................................................................................................5 3.3 Summary .................................................................................................................................5 4.0 RESEARCH DESIGN: ......................................................................................................................6 4.1 Exploratory Research ...............................................
Words: 2122 - Pages: 9
...Para os meus pais, porque "o valor das coisas não está no tempo que elas duram, mas na intensidade com que acontecem. Por isso existem momentos inesquecíveis, coisas inexplicáveis e pessoas incomparáveis" como vocês! Obrigado por tudo, Filipe Abstract The Retail Banking Industry has been severely affected by fraud over the past few years. Indeed, despite all the research and systems available, fraudsters have been able to outsmart and deceive the banks and their customers. With this in mind, we intend to introduce a novel and multi-purpose technology known as Stream Computing, as the basis for a Fraud Detection solution. Indeed, we believe that this architecture will stimulate research, and more importantly organizations, to invest in Analytics and Statistical Fraud-Scoring to be used in conjunction with the already in-place preventive techniques. Therefore, in this research we explore different strategies to build a Streambased Fraud Detection solution, using advanced Data Mining Algorithms and Statistical Analysis, and show how they lead to increased accuracy in the detection of fraud by at least 78% in our reference dataset. We also discuss how a combination of these strategies can be embedded in a Stream-based application to detect fraud in real-time. From this perspective, our experiments lead to an average processing time of 111,702ms per transaction, while strategies to further improve the performance are discussed. Keywords: Fraud Detection, Stream Computing, Real-Time...
Words: 56858 - Pages: 228
...Para os meus pais, porque "o valor das coisas não está no tempo que elas duram, mas na intensidade com que acontecem. Por isso existem momentos inesquecíveis, coisas inexplicáveis e pessoas incomparáveis" como vocês! Obrigado por tudo, Filipe Abstract The Retail Banking Industry has been severely affected by fraud over the past few years. Indeed, despite all the research and systems available, fraudsters have been able to outsmart and deceive the banks and their customers. With this in mind, we intend to introduce a novel and multi-purpose technology known as Stream Computing, as the basis for a Fraud Detection solution. Indeed, we believe that this architecture will stimulate research, and more importantly organizations, to invest in Analytics and Statistical Fraud-Scoring to be used in conjunction with the already in-place preventive techniques. Therefore, in this research we explore different strategies to build a Streambased Fraud Detection solution, using advanced Data Mining Algorithms and Statistical Analysis, and show how they lead to increased accuracy in the detection of fraud by at least 78% in our reference dataset. We also discuss how a combination of these strategies can be embedded in a Stream-based application to detect fraud in real-time. From this perspective, our experiments lead to an average processing time of 111,702ms per transaction, while strategies to further improve the performance are discussed. Keywords: Fraud Detection, Stream Computing, Real-Time...
Words: 56858 - Pages: 228
...Southern Cross University ePublications@SCU Theses 2004 The contribution of business/management education, to small enterprise solvency Peter Ellis Southern Cross University, PeterEllis@YSP.com.au Suggested Citation Ellis, P 2004, 'The contribution of business/management education, to small enterprise solvency', DBA thesis, Southern Cross University, Lismore, NSW. Copyright P Ellis 2004 For further information about this thesis Peter Ellis can be contacted at peterellis@ysp.com.au ePublications@SCU is an electronic repository administered by Southern Cross University Library. Its goal is to capture and preserve the intellectual output of Southern Cross University authors and researchers, and to increase visibility and impact through open access to researchers around the world. For further information please contact epubs@scu.edu.au. Southern Cross University Doctor of Business Administration The contribution of business/management education, to small enterprise solvency Peter Ellis Submitted to Graduate College of Management Southern Cross University, in partial fulfilment of the Degree of Doctor of Business Administration. 2004 Copyright “The contribution of business/management education, to small enterprise solvency.” Copyright © 2004 by Dr Peter Ellis, who reserves all rights and asserts his right under the Copyright, Design and Patents Act 1988, to be identified as the author of this work. No part of this work may be used or reproduced...
Words: 96678 - Pages: 387
