Some computational aspects of a distance—based model for prediction

Cuadras, Carles M.; Areans, C.; Fortiana, Josep

doi:10.1080/03610919608813332

Cited by 42 publications

(31 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Bereziki, polinomio ortogonalen gaineko erregresioaren baliokidea da [9] baldin eta erabilitako distantzia δ ij = (z i − z j ) 1/2 bada p = 1 izanik. [3,10] lanetan ikus daitezke emaitza osagarriak. Horretaz gain, aldagaien aukeraketarako F testaren parekoa den distantzietan oinarritutako testa [11] proposatu zen.…”

Section: Erregresioaunclassified

Estatistika metodoak distantzietan oinarrituriko ikuspegitik

Irigoien¹,

Arenas²,

Ferreiro³

et al. 2015

EKAIA

View full text Add to dashboard Cite

Laburpena: Lan honetan analisi anizkoitzaren baitan biltzen diren distantzietan oinarrituriko hainbat metodoren berrikusketa egin da. Oinarrizko kontzeptuak aurkeztu dira lehenik, ondoren metodoen funtsa laburki azaldu ahal izateko. Zehazki, erregresioa, diskriminazio-analisia, cluster-analisia, tipikotasuna eta sakonera aztertzen dituzten metodoak bildu ditugu. Metodologia honek berezitasun nabarmena du, edozein datu motaren gainean aplikagarria baita, datuek erakusten duten banaketa zein den ezagutu beharrik gabe. Azkenik, metodo hauen erabilgarritasuna erakusteko, funtzio-datu errealen gainean aplikatu eta lortutako emaitzak azaldu dira.Abstract: A revision of some distance-based methods is presented within the Multivariate Analysis. First, basic concepts are introduced so that those as regression, discrimination, clustering, typicality and depth methods are briefly explained. The main characteristic of distance based methods is that they can be applied to any type of mixed data, where a known distribution of data is not necessary. Finally, these methods have been applied to a real functional data set and the obtained results are shown. SARRERAEdozein azterketaren helburu diren aleen inguruan bildutako informazio ordenatuari datuak esaten diogu eta estatistikako metodoak datu horietatik ezagutza ateratzea ahalbidetzen duten matematikako teknikak dira. Estatistikako metodo batzuek alea × aldagaiak ikuspegian oinarritzen dira eta ondorioz datu matrizetik abiatzen dira. Beste metodo batzuek, aldiz, aleen arteko berdintasunak/desberdintasunak direlakoak aztertzean datza;

show abstract

Section: Erregresioaunclassified

Estatistika metodoak distantzietan oinarrituriko ikuspegitik

Irigoien¹,

Arenas²,

Ferreiro³

et al. 2015

EKAIA

View full text Add to dashboard Cite

show abstract

“…2.1 Distance-based linear model: definition and results DB-LM was introduced by Cuadras (1989) and has been developed in Cuadras and Arenas (1990), Cuadras et al (1996), , Esteve et al (2009) and Boj et al (2010). Here we recall its main concepts, as given in these articles, where the reader is referred to for more details and proofs.…”

Section: Review Of Db-lm and Glmmentioning

confidence: 99%

“…Based on the metric version of MDS, Cuadras (1989) introduced the distance-based linear model (DB-LM), later extended by Cuadras and Arenas (1990) and Cuadras et al (1996). They assume that for a set of individuals an interindividual distances matrix is available, as well as the value of a continuous response variable for each individual.…”

Section: Introductionmentioning

confidence: 99%

Global and local distance-based generalized linear models

Val

Mestres²,

Delicado

et al. 2015

TEST

View full text Add to dashboard Cite

This paper introduces local distance-based generalized linear models. These models extend (weighted) distance-based linear models firstly to the generalized linear model framework. Then a nonparametric version of these models is proposed by means of local fitting. Distances between individuals are the only predictor information needed to fit these models. Therefore they are applicable, among others, to mixed (qualitative and quantitative) explanatory variables or when the regressor is of functional type. An implementation is provided by the R package dbstats, which also implements other distancebased prediction methods. Supplementary material for this article is available online, which reproduces all the results of this article.

show abstract

“…For instance, Cuadras et al (1996) use the Principal Coordinates Euclidean configuration, X = U ·Λ, obtained from the eigendecomposition G = U · Λ 2 · U ′ . This version of DBR amounts to performing a Principal Components Regression (PCR) (see, e.g., Jolliffe 2002).…”

Section: Distance Based Regressionmentioning

confidence: 99%

“…Distance-Based Regression (DBR) (see Cuadras 1989, Cuadras & Arenas 1990, Cuadras et al 1996) is a method for predicting a numerical response y from a set z of both numerical and categorical predictors. The name of the procedure originates in the fact that it involves a metric in the space of predictors, d( · , · ), which must be Euclidean in the sense of Multidimensional Scaling (see Section 2).…”

Section: Introductionmentioning

confidence: 99%

Implementing PLS for distance-based regression: computational issues

Val

Grané

Fortiana

et al. 2007

Computational Statistics

View full text Add to dashboard Cite

Distance-based regression allows for a neat implementation of the Partial Least Squares recurrence. In this paper we address practical issues arising when dealing with moderately large datasets (n ~ 10 4 ) such as those typical of automobile insurance premium calculations.Keywords: Distance-based prediction; PLS regression; Large datasets; Automobile insurance data.(1) Dept. SummaryDistance-based regression allows for a neat implementation of the Partial Least Squares recurrence. In this paper we address practical issues arising when dealing with moderately large datasets (n ∼ 10 4 ) such as those typical of automobile insurance premium calculations.

show abstract

Some computational aspects of a distance—based model for prediction

Cited by 42 publications

References 16 publications

Estatistika metodoak distantzietan oinarrituriko ikuspegitik

Estatistika metodoak distantzietan oinarrituriko ikuspegitik

Global and local distance-based generalized linear models

Implementing PLS for distance-based regression: computational issues

Contact Info

Product

Resources

About