Modeling actuarial data with a composite lognormal-Pareto model

By Kahadawala Cooray, Malwane M. A. Ananda. 2005

240 citations recorded by the source corpus.

Subfields: Demography, Finance, Management Science and Operations Research

View Paper: https://www.semanticscholar.org/paper/f384591bfe15d9c067252eb7145f68d808dabff9

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Demography

In Finance

In Management Science and Operations Research

Citations in the corpus

Modeling Severity and Measuring Tail Risk of Norwegian Fire Claims (2016, background): “, McNeil, et al. 2005). Over the years, a number of authors have persuasively argued that the choice of GPD is natural and supported by rigorous theorems from extreme value theory, which is certainly true. In practice, however, implementation of this model is not as “smooth” as one would hope. Indee”
Skew mixture models for loss distributions: a Bayesian approach (2012, methodology): “Within the general framework of mixture models, relevant contributions to model the severity distributions have been proposed by Cooray and Amanda (2005), Scollnik (2007), who applied several composite log-Normal-Pareto models, and by Frigessi et al. (2002) who consider a dynamic mixture approach.”
The Heavy-Tailed Exponential Distribution: Risk Measures, Estimation, and Application to Actuarial Data (2020, background): “The insurance data sets are usually positive (Klugman et al. [1]), unimodal shaped (Cooray and Ananda [2]), right-skewed (Lane [3]), and with heavy tails (Ibragimov and Prokhorov [4]).”
The Extended Log-Logistic Distribution: Inference and Actuarial Applications (2021, background): “The insurance data are usually unimodal [1], right-skewed [2], positive [3], have a heavy-tailed density [4], and have a unimodal hump shape [5].”
Composite Lognormal–Pareto model with random threshold (2011, methodology): “However, the proposal made by Cooray & Ananda (2005) has been amended by Scollnik (2007) who pointed out that this model fixes the proportion of large claims, which appears very restrictive.”
Modeling loss data using mixtures of distributions (2016, background): “Coorey and Ananda (2005) introduced a composite Log-normal-Pareto model. Their model is composed of a Lognormal density up to an unknown threshold and a two-parameter Pareto beyond the threshold. This composition is justified because the Pareto model fits well the upper long tail of the distribution”
Modeling loss data using composite models (2015, methodology): “The data is highly skewed and have been applied in many papers, including Cooray and Ananda (2005), Scollnik (2007), Scollnik and Sun (2012) and Nadarajah and Bakar (2014) to model insurance losses using compositemodels.”
Second-order refined peaks-over-threshold modelling for heavy-tailed distributions (2009, methodology): “In an actuarial context, the relevance of using more elaborate models has already been discussed for instance in Frigessi et al. (2002) and Cooray and Ananda (2005).”
A New Extension of Generalized Exponential Distribution with Application to Ozone Data (2017, methodology): “…and Marshall and Olkin (1997); beta-generated method by Eugene, Lee, and Famoye (2002); transformed-transformer method (T-X family) by Alzaatreh et al. (2013); composite method by Cooray and Ananda (2005); inverse probability integral transformation method by Ferreira and Steel (2006) and so on.”