Multivariate measurement error models using finite mixtures of skew-Student <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" altimg="si56.gif" display="inline" overflow="scroll"><mml:mi>t</mml:mi></mml:math> distributions

Cabral, Celso Rômulo Barbosa; Lachos, Víctor H.; Zeller, Camila Borelli

doi:10.1016/j.jmva.2013.10.017

Cited by 23 publications

(12 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These criteria quantitatively evaluate the suitability of a statistical distribution to model a data set by penalizing the number of parameters in each distribution to avoid data overfitting. The three information criteria considered here, namely the Akaike information criterion (AIC), Bayesian information criterion (BIC), and efficient determination criterion (EDC), have the general form of [21]…”

Section: B the Information Criteriamentioning

confidence: 99%

See 1 more Smart Citation

Statistical Modeling of Brazilian In-Home PLC Channel Features

Oliveira¹,

Picorone

Zeller

et al. 2019

JCIS

View full text Add to dashboard Cite

This paper deals with the statistical modeling of key features of power line communication (PLC) channels that are necessary for designing data communication systems that operate over theses channels. The key features are average channel attenuation, root mean squared delay spread, coherence bandwidth and coherence time. All these features were estimated from in-home PLC channels measured in seven distinct Brazilian residences. Assuming that each feature is a random variable, four criteria (i.e., maximum likelihood estimate and three different information criteria) are used to select the statistical distribution that fits best to the data. The symmetry and asymmetry of the histogram associated with each feature is pointed out. The reported results focus on three frequency bands, namely: 1.7-30 MHz, 1.7-50 MHz and 1.7-100 MHz, which are in accordance with the standards used in Europe, North America and Brazil. The values obtained to Brazilian PLC channel features are different from those related to US and Europe channels reported in literature. Thus, the presented statistical models constitute an important tool to better design practical PLC systems that are suitable for Brazilian and in-home electric power grids.

show abstract

Section: B the Information Criteriamentioning

confidence: 99%

“…• The choice of the best statistical distribution is based on the log-likelihood function and three distinct information criteria, namely Akaike information criterion (AIC), Bayesian information criterion (BIC), and efficient determination criterion (EDC) [21]. • The analysis comprises three distinct frequency bands:…”

Section: Introductionmentioning

confidence: 99%

Statistical Modeling of Brazilian In-Home PLC Channel Features

Oliveira¹,

Picorone

Zeller

et al. 2019

JCIS

View full text Add to dashboard Cite

show abstract

“…Based on Cabral et al, Akaike information criterion (AIC), Bayesian information criterion (BIC), efficient determination criterion (EDC), and log‐likelihood function are used as evaluation criteria to choose the best model of the nSNR coherence bandwidth. For the log‐likelihood function, the distribution that offers the best statistical model is the one with maximum log‐likelihood value.…”

Section: Statistical Analysis and Modelingmentioning

confidence: 99%

Statistical analysis and modeling of a novel parameter for resource allocation in multicarrier PLC systems

Colen

Oliveira

Zeller

et al. 2017

Trans Emerging Tel Tech

View full text Add to dashboard Cite

This work outlines a novel parameter that is capable of characterizing the flatness of the normalized signal-to-noise ratio for resource allocation purposes in power line communication (PLC) systems, which are based on multicarrier modulation schemes. This parameter, named normalized signal-to-noise ratio coherence bandwidth, can be used for grouping adjacent subchannels therefore reducing the computational complexity of resource allocation techniques. Additionally, we discuss a comparative analysis between the proposed parameter and the coherence bandwidth for resource allocation purposes, which leads us to conclude that the former is more appropriate than the latter for PLC systems. Moreover, we show a statistical analysis of the proposed parameter and model it with different distributions based on noise measurements and channel estimates of broadband in-home, outdoor, and hybrid PLC-wireless environments. On the basis of data sets obtained from measurement campaigns covering the frequency bands of 1.7 to 30, 1.7 to 50, and 1.7 to 100 MHz, we show that the best distributions for modeling the proposed parameter are inverse Gaussian, Nakagami, gamma, and Gaussian mixture, depending on the type of the channel (in home, outdoor, and hybrid PLC wireless) and the frequency band.

show abstract

“…The mixsmsn package in R (Cabral et al, 2014;Prates et al, 2013b) considers mixtures of Gaussian, t, skew-t, skew normal, skew contaminated normal and skew slash contaminated normal models. The optimal candidate among the mixsmsn models, as determined by BIC, is a G = 2 mixture of skew-t distributions with unequal covariance structure across groups and BIC = −2313.0.…”

Section: Old Faithful Datamentioning

confidence: 99%