Discriminant Function Analysis (Statistical Associates Blue Book Series 27)

Discriminant Function Analysis (Statistical Associates Blue Book Series 27)

DISCRIMINANT FUNCTION ANALYSIS <br /><br />Discriminant function analysis, also known as discriminant analysis or simply DA, is used to classify cases into the values of a categorical dependent, usually a dichotomy. If discriminant function analysis is effective for a set of data, the classification table of correct and incorrect estimates will yield a high percentage correct. Discriminant function analysis is found in SPSS under Analyze>Classify>Discriminant. If the specified grouping variable has two categories, the procedure is considered "discriminant analysis" (DA). If there are more than two categories the procedure is considered "multiple discriminant analysis" (MDA). <br /><br />The full content is now available from Statistical Associates Publishers. Http://www.statisticalassociates.com<br /><br />Below is the unformatted table of contents. <br /><br />Table of Contents<br /><br />Overview6<br />Key Terms and Concepts7<br />Variables7<br />Discriminant functions7<br />Pairwise group comparisons8<br />Output statistics8<br />Examples9<br />SPSS user interface9<br />The "Statistics" button10<br />The "Classify" button10<br />The "Save" button13<br />The "Bootstrap" button13<br />The "Method" button14<br /> SPSS Statistical output for two-group DA16<br />The "Analysis Case Processing Summary" table16<br />The "Group Statistics" table16<br />The "Tests of Equality of Group Means" table16<br />The "Pooled Within-Group Matrices" and "Covariance Matrices" tables.18<br />The "Box's Test of Equality of Covariance Matrices" tables18<br />The "Eigenvalues" table19<br />The "Wilks' Lambda" table21<br />The "Standardized Canonical Discriminant Function Coefficients" table21<br />The "Structure Matrix" table23<br />The "Canonical Discriminant Functions Coefficients" table23<br />The "Functions at Group Centroids" table24<br />The "Classification Processing Summary" table24<br />The "Prior Probabilities for Groups" table25<br />The "Classification Function Coefficients" table25<br />The "Casewise Statistics" table26<br />Separate-groups graphs of canonical discriminant functions27<br />The "Classification Results" table27<br />SPSS Statistical output for three-group MDA28<br />Overview and example28<br />MDA and DA similarities28<br />The "Eigenvalues" table29<br />The "Wilks' Lambda" table29<br />The "Structure Matrix" table30<br />The "Territorial Map"31<br />Combined-groups plot34<br />Separate-groups plots34<br />SPSS Statistical output for stepwise discriminant analysis35<br />Overview35<br />Example35<br />Stepwise discriminant analysis in SPSS36<br />Assumptions41<br />Proper specification41<br />True categorical dependent variables41<br />Independence41<br />No lopsided splits41<br />Adequate sample size41<br />Interval data42<br />Variance42<br />Random error42<br />Homogeneity of variances (homoscedasticity)42<br />Homogeneity of covariances/correlations42<br />Absence of perfect multicollinearity43<br />Low multicollinearity of the independents43<br />Linearity43<br />Additivity43<br />Multivariate normality43<br />Frequently Asked Questions44<br />Isn't discriminant analysis the same as cluster analysis?44<br />When does the discriminant function have no constant term?44<br />How important is it that the assumptions of homogeneity of variances and of multivariate normal distribution be met?44<br />In DA, how can you assess the relative importance of the discriminating variables?44<br />Dummy variables45<br />In DA, how can you assess the importance of a set of discriminating variables over and above a set of control variables? (What is sequential discriminant analysis?)45<br />What is the maximum likelihood estimation method in discriminant analysis (logistic discriminate function analysis)?45<br />What are Fisher's linear discriminant functions?46<br />I have heard DA is related to MANCOVA. How so?46<br />How does MDA work?46<br />How can I tell if MDA worked?46<br />For any given MDA example, how many discriminant functions will there be, and how can I tell if each is significant?47<br />What are Mahalonobis distances?47<br />How are the multiple discriminant scores on a single case interpreted in MDA?47<br />And 5 more pages of topics on MDA.<br />Coverage: SPSS<br />Pagecount: 52