Papers by Sotirios Loukas
Estimation for the bivariate negative binomial distribution
Journal of Statistical Computation and Simulation, 1985
ML estimation in the bivariate “short” distributions by the EM algorithm
The maximum likelihood estimation of parameters of two bivariate ‘Short’ distributions, introduce... more The maximum likelihood estimation of parameters of two bivariate ‘Short’ distributions, introduced by H. Papageorgiou [Commun. Stat., Theory Methods 15, 893-905 (1986; Zbl 0613.62021)], is considered by applying the EM algorithm. The observed Fisher informations are derived and numerical examples based on real data are presented where the convergence of the EM algorithm is accelerated substantially by the methods of conjugate gradients and T. A. Louis [J. R. Stat. Soc., Ser. B 44, 226-233 (1982; Zbl 0488.62018)].
New ineq u a lities are o b ta in e d for S p e a rm a n 's fo o tru le , its asso ciate d sum o ... more New ineq u a lities are o b ta in e d for S p e a rm a n 's fo o tru le , its asso ciate d sum o f th e ab so lu te values o f ran k differen ces, th e n u m b e r o f d isc o rd a n t pairs asso ciated w ith K en d all's ta u a n d th e c o efficie n t o f c o n c o rd a n c e o f several rankings w hen o n e or m ore individuals are m issing or a d d ed to th e d ata.
Statistics & Probability Letters, 1991
The inequality on the sum of the absolute values of rank differences associated with Spearman's f... more The inequality on the sum of the absolute values of rank differences associated with Spearman's footrule, which was established in Papaioannou and Loukas , is shown to hold when ties are present in both rankings. The same result is established numerically for the sum of squares of rank differences associated with Spearman's rank correlation coefficient, when ties are present in both rankings.
The use of soft and flexible guidewires in the treatment of chronic total coronary occlusions by activated guidewire angioplasty
The British journal of radiology, 1999
Activated guidewire angioplasty (AGA) is a new technique which has been designed to assist in ang... more Activated guidewire angioplasty (AGA) is a new technique which has been designed to assist in angioplasty of total occlusions. The purpose of this study was to determine the safety and efficacy of using flexible relatively soft guidewires (floppy wires) in conjunction with this technique and also to determine the predictors of lesion crossing and final success by this technique in patients with chronic total coronary occlusions. 73 patients with 73 chronic total coronary occlusions in whom coronary angioplasty using conventional techniques had failed were treated with AGA using floppy guidewires. The success of crossing these lesions was 65.7% (48/73) resulting in a final angioplasty success of 56.1% (41/73). Angioplasty success was reduced compared with crossing success in seven arteries in which complications occurred during balloon angioplasty. Multiple stepwise logistic regression analysis identified the location of the occlusion (right coronary artery, p = 0.005) as independent...
Relatively low red cell folate levels and acute coronary syndromes
Coronary Artery Disease, 2001
Low folate levels are related to increased risk for coronary artery disease in humans, while expe... more Low folate levels are related to increased risk for coronary artery disease in humans, while experimental work has shown that folate deficiency is thrombogenic. We hypothesized that relatively low folate levels are related to the development of acute coronary syndromes in patients with previously stable coronary artery disease. One hundred and forty-one men were studied: 53 consecutive patients with acute coronary syndromes, 41 with stable coronary artery disease and 47 control participants. Known clinical and lipid risk factors were identified in all subjects and in addition plasma B12, plasma and red cell folate levels were measured. Red cell folate levels were significantly lower in patients with acute coronary syndromes (510+/-178 nmol/l) than in both stable coronary artery disease patients (638+/-264 nmol/l, P< 0.005) and controls (615+/-193 nmol/l, P< 0.05 respectively). Plasma folate and B12 levels were similar in all three groups. Multiple logistic regression analysis identified red cell folate levels as the only independent predictor of acute coronary events in the whole population of patients with known coronary artery disease and in the subgroup of non-smokers (P=0.010 and P=0.031). The present study suggests that relatively low red cell folate levels are associated with acute coronary syndromes and are an independent predictor of acute coronary events.
Annals of the Institute of Statistical Mathematics, Feb 1, 1999
Statistics Probability Letters, Jul 1, 2005
Various statistical properties and reliability aspects of a two-parameter distribution with decre... more Various statistical properties and reliability aspects of a two-parameter distribution with decreasing and increasing failure rate are explored; the model includes the exponential-geometric distribution . Biometrika 84, 641-652; Adamidis and Loukas, 1998. Statist. Probab. Lett. 39, 35-42] as a special case. Characterizations are given and the estimation of parameters is studied by the method of maximum likelihood. An EM algorithm [Dempster et al., 1977. J. R. Statist. Soc. B. 39, 1-38] is proposed for computing the estimates, and expressions for their asymptotic variances and covariances are derived. Numerical examples based on real data are included.
Statistics Probability Letters, Jul 1, 1998
A two-parameter distribution with decreasing failure rate is introduced. Various properties are d... more A two-parameter distribution with decreasing failure rate is introduced. Various properties are discussed and the estimation of parameters is studied by the method of maximum likelihood. The estimates are attained by the EM algorithm and expressions for their asymptotic variances and covariances are obtained. Numerical examples based on real data are presented. ~)
Communications in Statistics Theory and Methods, Jun 27, 2007
Communications in Statistics is a multi-part journal.
Applications of Mathematics, 1991
A three-parameter lifetime distribution
ABSTRACT A three-parameter distribution is introduced, as an extension of the lifetime model of C... more ABSTRACT A three-parameter distribution is introduced, as an extension of the lifetime model of Chen [8]; various properties are explored and its usefulness in modeling real data is demonsrated using maximum likelihood estimates.
Communications in Statistics-theory and Methods, 2012
In this article four derivations are presented, for an absolutely continuous bivariate extension ... more In this article four derivations are presented, for an absolutely continuous bivariate extension of the Extended Exponential-Geometric distribution (EEG) introduced by Adamidis et al. (2005). Three of these derivations are based on “shock models” and one is based on the assumption of a two component system working in a varying environment. Marginal and conditional distributions are obtained and their corresponding
Fast general methods for generating univariate discrete random variables require the preliminary ... more Fast general methods for generating univariate discrete random variables require the preliminary setting-up of tables. Actual generation then involves a table look-up procedure initiated by a generated pseudo-random uniform variable. Two such methods were introduced by Marsaglia (1963) and one by Walker (1974). . KEY WORDS. Computer generation, bivariate discrete distributions, alias generation method, non-sequential search procedures. . C opyright © 1981 b y D. R eid el Publishing Company.
Following a conjecture in Kemp and Loukas (J. Roy. Statist. Soc. A 1 4 1 , in press) a new algori... more Following a conjecture in Kemp and Loukas (J. Roy. Statist. Soc. A 1 4 1 , in press) a new algorithm for generating bivariate discrete random variables has been implemented. Like other inverse-interpolation methods considered in the paper, it involves computing an array of probabil ities and accumulating them, but the probabilities are ordered to form a monotonically decreasing sequence before accumulation. It requires more storage than the other methods and takes much more time to set up the flook-up1 table of accumulated ordered probabilities. However the time then taken to generate a random vector is quite small and is easily further reduced by incorporating the 'index table1 method described by Chen and Asau (AIIE T rans. J5, 163). The algorithm has been tested on the bivariate Poisson and Hermite distributions used in Kemp and Loukas (l oc. c i t .) and its gener ation times compared favourably with those of the fastest algorithms described there. INTRODUCTION
AM All use subject to JSTOR Terms and Conditions The Computer Generation of Bivariate Discrete Random Variables
about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, an... more about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. . Wiley and Royal Statistical Society are collaborating with JSTOR to digitize, preserve and extend access to SUMMARY Two different types of procedure for generating samples from bivariate discrete distributions are studied. The first ones are general in character and based on inverse interpolation in arrays of accumulated probabilities. The others are specific to individual distributions, are based on the stochastic structure of a distribution and combine univariate random variables in various ways. Timings are given for all the methods when used for bivariate Poisson and bivariate Hermite distributions. One of the general...
Journal of the Royal Statistical Society Series A (General)
Two different types of procedure for generating samples from bivariate discrete distributions are... more Two different types of procedure for generating samples from bivariate discrete distributions are studied. The first ones are general in character and based on inverse interpolation in arrays of accumulated probabilities. The others are specific to individual distributions, are based on the stochastic structure of a distribution and combine univariate random variables in various ways. Timings are given for all the methods when used for bivariate Poisson and bivariate Hermite distributions. One of the general methods is recommended if storage requirements are not critical, but the stochastic methods are attractive where very little storage is available, e.g. programmable pocket calculators.
On the gamma frailty model
One of the ways of assessing the impact of heterogeneity in mortality studies is via the concept ... more One of the ways of assessing the impact of heterogeneity in mortality studies is via the concept of “frailty”, introduced by J. W. Vaupel et al. [Demography 16, 439–454 (1979)]. When the multiplicative frailty model is under consideration [see, for instance, J. W. Vaupel, Kindred lifetimes: frailty models in population genetics. J. Adams et al. (eds.), Convergent Issues in Genetics and Demography, 156-170 (1990)], the assumption of a gamma distributed frailty leads to the so-called gamma frailty model. This paper is devoted to exploiting some aspects of its relevant distribution theory; failure rate characterizations are obtained and bounds on the survival function are constructed. Moreover, it is shown that the model can serve as a method for constructing lifetime models or extending existing ones [by adding a parameter in the sense of A. W. Marshall and I. Olkin, Biometrika 84, No. 3, 641–652 (1997), correction ibid. 92, No. 2, 505 (2005; Zbl 0888.62012)].
International Journal of Quality, Statistics, and Reliability, 2012
A four-parameter family of Weibull distributions is introduced, as an example of a more general c... more A four-parameter family of Weibull distributions is introduced, as an example of a more general class created along the lines of Marshall and Olkin, 1997. Various properties of the distribution are explored and its usefulness in modelling real data is demonstrated using maximum likelihood estimates.
Uploads
Papers by Sotirios Loukas