In this work we deal with classifying mixed type data using a hierar-chical approach based on Forward Search. The identification of groups is basedon the identification of similar trajectories and then linked to very intuitive two-dimensional maps. The proposed algorithm can implement different measures formixed type data, such as Gower or Related Mertric Scaling. The algorithm will beapplied to data related to a large set of countries. In economics it is a shared idea thata country’s economic performance, but also more or less sustainable developmentpolicies depend on their deep institutional characteristics. These characteristics aredescribed by indicators of mixed nature.
Nel presente lavoro ci occupiamo di classificare dati misti utilizzando unapproccio gerarchico basato sulla Forward Search. L’individuazione dei gruppi sibasa sull’identificazione di traiettorie simili e poi collegate a mappe bidimension-ali molto ituitive. L’algoritmo proposto pu implementare diverse misure per datimisti, per esempio Gower o Related Mertric Scaling. L’algoritmo sar applicato adati relativi a un ampio insieme di paesi. In economia una idea condivisa che leperformance economiche, ma anche le politiche di sviluppo pi o meno sostenibile,di un paese dipendono dalle loro profonde caratteristiche istituzionali. Tali carat-terisitche sono descritte da indicatori di natura mista.
Robust multivariate analysis for mixed-type data = Analisi multivariata robusta per dati misti / A. Grané, S. Salini - In: Statistical Methods for Service Quality Evaluation / [a cura di] M. Bini, P. Amenta, A. D'Ambra, I. Camminatiello. - Prima edizione. - [s.l] : Cuzzolin, 2019 Jul. - ISBN 9788886638654. - pp. 82-86 (( Intervento presentato al 9. convegno Statistical evaluation systems at 360°: techniques, technologies and new frontiers tenutosi a Roma nel 2019.
Robust multivariate analysis for mixed-type data = Analisi multivariata robusta per dati misti
S. Salini
Secondo
2019
Abstract
In this work we deal with classifying mixed type data using a hierar-chical approach based on Forward Search. The identification of groups is basedon the identification of similar trajectories and then linked to very intuitive two-dimensional maps. The proposed algorithm can implement different measures formixed type data, such as Gower or Related Mertric Scaling. The algorithm will beapplied to data related to a large set of countries. In economics it is a shared idea thata country’s economic performance, but also more or less sustainable developmentpolicies depend on their deep institutional characteristics. These characteristics aredescribed by indicators of mixed nature.File | Dimensione | Formato | |
---|---|---|---|
IES 2019 Book of Short Papers.pdf
accesso aperto
Tipologia:
Publisher's version/PDF
Dimensione
1.17 MB
Formato
Adobe PDF
|
1.17 MB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.