Andrea Cerioli – författare

Why We Wrote This Book This book is about using graphs to explore and model continuous multi­ variate data. Such data are often modelled using the multivariate normal distribution and, indeed, there is a literatme of weighty statistical tomes presenting the mathematical theory of this activity. Our book is very dif­ ferent. Although we use the methods described in these books, we focus on ways of exploring whether the data do indeed have a normal distribution. We emphasize outlier detection, transformations to normality and the de­ tection of clusters and unsuspected influential subsets. We then quantify the effect of these departures from normality on procedures such as dis­ crimination and duster analysis. The normal distribution is central to our book because, subject to our exploration of departures, it provides useful models for many sets of data. However, the standard estimates of the parameters, especially the covari­ ance matrix of the observations, are highly sensitive to the presence of outliers. This is both a blessing and a curse. It is a blessing because, if we estimate the parameters with the outliers excluded, their effect is appre­ ciable and apparent if we then include them for estimation. It is however a curse because it can be hard to detect which observations are outliers. We use the forward search for this purpose.

Why We Wrote This Book This book is about using graphs to explore and model continuous multi­ variate data. Such data are often modelled using the multivariate normal distribution and, indeed, there is a literatme of weighty statistical tomes presenting the mathematical theory of this activity. Our book is very dif­ ferent. Although we use the methods described in these books, we focus on ways of exploring whether the data do indeed have a normal distribution. We emphasize outlier detection, transformations to normality and the de­ tection of clusters and unsuspected influential subsets. We then quantify the effect of these departures from normality on procedures such as dis­ crimination and duster analysis. The normal distribution is central to our book because, subject to our exploration of departures, it provides useful models for many sets of data. However, the standard estimates of the parameters, especially the covari­ ance matrix of the observations, are highly sensitive to the presence of outliers. This is both a blessing and a curse. It is a blessing because, if we estimate the parameters with the outliers excluded, their effect is appre­ ciable and apparent if we then include them for estimation. It is however a curse because it can be hard to detect which observations are outliers. We use the forward search for this purpose.

Why We Wrote This Book This book is about using graphs to explore and model continuous multi­ variate data. Such data are often modelled using the multivariate normal distribution and, indeed, there is a literatme of weighty statistical tomes presenting the mathematical theory of this activity. Our book is very dif­ ferent. Although we use the methods described in these books, we focus on ways of exploring whether the data do indeed have a normal distribution. We emphasize outlier detection, transformations to normality and the de­ tection of clusters and unsuspected influential subsets. We then quantify the effect of these departures from normality on procedures such as dis­ crimination and duster analysis. The normal distribution is central to our book because, subject to our exploration of departures, it provides useful models for many sets of data. However, the standard estimates of the parameters, especially the covari­ ance matrix of the observations, are highly sensitive to the presence of outliers. This is both a blessing and a curse. It is a blessing because, if we estimate the parameters with the outliers excluded, their effect is appre­ ciable and apparent if we then include them for estimation. It is however a curse because it can be hard to detect which observations are outliers. We use the forward search for this purpose.

This volume contains revised versions of selected papers presented at the biennial meeting of the Classification and Data Analysis Group (CLADAG) of the Italian Statistical Society, which was held in Parma, June 6-8, 2005. Sergio Zani chaired the Scientific Programme Committee and Andrea Cerioli chaired the Local Organizing Committee. The scientific programme of the conference included 127 papers, 42 in spe­ cialized sessions, 68 in contributed paper sessions and 17 in poster sessions. Moreover, it was possible to recruit five notable and internationally renowned invited speakers (including the 2004-2005 President of the International Fed­ eration of Classification Societies) for plenary talks on their current research work. Among the specialized sessions, two were organized by Wolfgang Gaul with five talks by members of the GfKl (German Classification Society), and one by Jacqueline J. Meulman (Dutch/Flemish Classification Society). Thus, the conference provided a large number of scientists and experts from home and abroad with an attractive forum for discussion and mutual exchange of knowledge. The topics of all plenary and specialized sessions were chosen to fit, in the broadest possible sense, the mission of CLADAG, the aim of which is "to further methodological, computational and applied research within the fields of Classification, Data Analysis and Multivariate Statistics". A peer-review refereeing process led to the selection of 46 extended papers, which are contained in this book.