## COMPSTAT: Proceedings in Computational Statistics, 9th Symposium held at Dubrovnik, Yugoslavia, 1990Konstantin Momirovic, Vesna Mildner Although no-one is, probably, too enthused about the idea, it is a fact that the development of most empirical sciences to a great extent depends on the development of data analysis methods and techniques, which, due to the necessity of application of computers for that purpose, actually means that it practically depends on the advancement and orientation of computer statistics. Every other year the International Association for Statistical Computing sponsors the organizition of meetings of individual s professiona77y involved in computational statistics. Since these meetings attract professionals from allover the world, they are a good sample for the estimation of trends in this area which some believe is a statistics proper while others claim it is computer science. It seems, though, that an increasing number of colleagues treat it as an independent scientific or at least technical discipline. This volume contains six invited papers, 41 contributed papers and, finally, two papers which are, formally, software descriptions, but it was agreed by the Program Committee that they should be included in a separate section entitled "Notes about new developments in statistical software", due to their special significance for current trends in computational statistics. |

### What people are saying - Write a review

We haven't found any reviews in the usual places.

### Contents

3 | |

Ferligoj A 15 | 20 |

PROTOTYPING DYNAMIC GRAPHICS FUNCTIONS IN | 35 |

STATISTICAL KNOWLEDGE BASED SYSTEMS CRITICAL REMARKS | 48 |

A DISCRIMINANT ANALYSIS | 63 |

AN EXPERT SYSTEM STRATEGY FOR SELECTING INTERESTING RESULTS | 81 |

ON MODEL SEARCH METHODS | 100 |

SELECTING THE BEST SUBSET OF WARIABLES | 115 |

ALTERNATE FORMS OF GRAPHICAL MODELLING A COMPARISON | 133 |

SOFTWARE POTENTIAL | 151 |

### Other editions - View all

Compstat 1990 - Proceedings in Computational Statistics Konstantin Momirovic,Vesna Mildner No preview available - 1990 |

### Common terms and phrases

algorithm analysis methods applied approach approximation Association for Statistical biplots BOJA bootstrap choice classification clustering coefficients Compstat concept considered contingency tables corresponding covariance matrix criterion data analysis data sets Database defined denote device driver distribution eigenvalues eigenvectors example expert system factor GENSTAT given GLIM graphical modelling Heidelberg for IASC hypergraph IASC International Association independence graph influence function interaction iteration jackknife language linear model log-linear models loglinear model M-estimators minimize multiple multivariate non-linear observations obtained optimal outliers parameters partition Physica-Verlag Heidelberg Principal Component Analysis PRLA probability problem procedure projection random RED(S regression representation resampling residuals robust estimate Robust Statistics rotation Rousseeuw scatterplot selection sequence sequential sampling plan simulated simulated annealing solution specification standard Statistical Computing statistical consultation statisticians stochastic structure subset techniques values variables variance vector WAMASTEX