Search Results - subject:"relative error"

An empirical study of hierarchical Bayes small area estimators using different priors for model variances

You, Yong - In: Statistics in transition : an international journal of … 24 (2023) 4, pp. 169-178

In this paper, we study hierarchical Bayes (HB) estimators based on different priors for small area estimation. In particular, we use inverse gamma and flat priors for variance components in the HB small area models of You and Chapman (2006) and You (2021). We evaluate the HB estimators through...

Persistent link: https://www.econbiz.de/10015115089

Composite Likelihood Inference by Nonparametric Saddlepoint Tests

Lunardon, Nicola; Ronchetti, Elvezio - Dipartimento di Scienze Economiche, Aziendali, … - 2013

the composite score functions. The proposed statistic is asymptotically chi-square distributed up to a relative error of …

Persistent link: https://www.econbiz.de/10010856295

Efficient simulation of tail probabilities of sums of correlated lognormals

Asmussen, Soren; Blanchet, Jose; Juneja, Sandeep; … - 2010

Persistent link: https://www.econbiz.de/10009448818

Improved algorithms for rare event simulation with heavy tails

Asmussen, Søren; Kroese, Dirk P. - 2006

the other uses conditional Monte Carlo conditioning upon Y1,..., Yn-1. Properties of the relative error of the estimators …, the conditional Monte Carlo estimator is the first heavy-tailed example of an estimator with bounded relative error …

Persistent link: https://www.econbiz.de/10009448797

Importancia relativa de los coeficientes y las transacciones de una estructura insumo-producto

Soto, Vicente Germán - In: Economía Mexicana NUEVA ÉPOCA X (2001) 2, pp. 325-359

To every economy, the problem set to find is to establish which sectors are most sensitive to changes in the final demand. In this paper we derive the mathematical conditions to obtain the relative importance of an input coefficients. The method is based on the impacts of gross products before...

Persistent link: https://www.econbiz.de/10008493895