Population Symbolic Covariance Matrices for Interval Data
Preprint
Abstract
Symbolic Data Analysis (SDA) is a relatively new field of statistics that extends classical data analysis by taking into account intrinsic data variability and structure. As SDA has been mainly approached from a sampling perspective, we introduce population formulations of the symbolic mean, variance, covariance, correlation, covariance matrix and correlation matrix for interval-valued symbolic variables, providing a theoretical framework that gives support to interval-valued SDA. Moreover, we provide an interpretation of the various definitions of covariance and correlation matrices according to the structure of micro-data, which allows selecting the model that best suits specific datasets. Our results are illustrated using two datasets. Specifically, we select the most appropriate model for each dataset using goodness-of-fit tests and quantile-quantile plots, and provide an explanation of the micro-data based on the covariance matrix.Keywords
All Related Versions
This publication has 18 references indexed in Scilit:
- Outlier detection in interval dataAdvances in Data Analysis and Classification, 2017
- Off the beaten track: A new linear model for interval dataEuropean Journal of Operational Research, 2017
- Kernelized inner product-based discriminant analysis for interval dataPattern Analysis and Applications, 2017
- Discriminant Analysis of Interval Data: An Assessment of Parametric and Distance-Based ApproachesJournal of Classification, 2015
- Symbolic Data Analysis: another look at the interaction of Data Mining and StatisticsWIREs Data Mining and Knowledge Discovery, 2014
- CIPCA: Complete-Information-based Principal Component Analysis for interval-valued dataNeurocomputing, 2012
- Likelihood functions and some maximum likelihood estimators for symbolic dataJournal of Statistical Planning and Inference, 2011
- Far beyond the classical data models: symbolic data analysisStatistical Analysis and Data Mining: An ASA Data Science Journal, 2011
- Forecasting models for interval-valued time seriesNeurocomputing, 2008
- Dynamic clustering for interval data based on L 2 distanceComputational Statistics, 2006