Dispersion modelling of mortality for both sexes with Tweedie distributions

Li, Jackie; Pitt, David; Li, Han

doi:10.1080/03461238.2021.1980430

Cited by 4 publications

(5 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The latter is used as the maximum lag. Unfortunately, in our case, there is no available asymptotic results for the behavior of the autocorrelation in (16), in particular, when the underlying random field is described using the AR-ARCH structure in Equation (7). Developing a specific test for this particular model is challenging but goes beyond the scope of the current paper.…”

Section: (I) Model Selectionmentioning

confidence: 96%

“…For instance, [15] proposed using the Poisson assumption to model the number of deaths and [7] further allowed for overdispersion. Recently, Reference [16] developed an additional dispersion sub-model alongside the mean model for a joint modeling of the mean and dispersion. This line of literature is totally missing.…”

Section: Classic Mortality Modelsmentioning

confidence: 99%

“…In fact, when considering larger sets, the number of models grows drastically. Indeed, when one considers the maximal sets V 1 and V 2 , as depicted in Figure 2 for instance, we should test 2 16 = 65,536 models, making this particular step very time-consuming. In fact, the selection procedure involves the optimization of the quasi-likelihood in order to estimate the parameters of each candidate model.…”

Section: (Iii) Bayesian Information Criterionmentioning

confidence: 99%

“…In Figure 4, we report the results of the selection procedure. A set of 2 16 models were investigated and the one maximizing the BIC criterion was selected. The parameters of the neighborhood with "-" do not belong to the optimal neighborhood.…”

Section: (I) Model Selectionmentioning

confidence: 99%

“…Furthermore, we want to ensure that the residuals do not exhibit any serial correlation, which may violate the underlying assumption of independence. In Figure A1, we plot the spatial autocorrelation of the standardized residuals, as implied by the selected model using the definition in Equation (16). None of the plots provide any evidence of a persistent autocorrelation.…”

Section: (I) Model Selectionmentioning

confidence: 99%

See 4 more Smart Citations

Optimal Neighborhood Selection for AR-ARCH Random Fields with Application to Mortality

Doukhan

Salhi

2021

Stats

View full text Add to dashboard Cite

This article proposes an optimal and robust methodology for model selection. The model of interest is a parsimonious alternative framework for modeling the stochastic dynamics of mortality improvement rates introduced recently in the literature. The approach models mortality improvements using a random field specification with a given causal structure instead of the commonly used factor-based decomposition framework. It captures some well-documented stylized facts of mortality behavior including: dependencies among adjacent cohorts, the cohort effects, cross-generation correlations, and the conditional heteroskedasticity of mortality. Such a class of models is a generalization of the now widely used AR-ARCH models for univariate processes. A the framework is general, it was investigated and illustrated a simple variant called the three-level memory model. However, it is not clear which is the best parameterization to use for specific mortality uses. In this paper, we investigate the optimal model choice and parameter selection among potential and candidate models. More formally, we propose a methodology well-suited to such a random field able to select thebest model in the sense that the model is not only correct but also most economical among all thecorrectmodels. Formally, we show that a criterion based on a penalization of the log-likelihood, e.g., the using of the Bayesian Information Criterion, is consistent. Finally, we investigate the methodology based on Monte-Carlo experiments as well as real-world datasets.

show abstract

Section: (I) Model Selectionmentioning

confidence: 96%

Section: Classic Mortality Modelsmentioning

confidence: 99%

Section: (Iii) Bayesian Information Criterionmentioning

confidence: 99%

Section: (I) Model Selectionmentioning

confidence: 99%

Section: (I) Model Selectionmentioning

confidence: 99%

See 3 more Smart Citations

Optimal Neighborhood Selection for AR-ARCH Random Fields with Application to Mortality

Doukhan

Salhi

2021

Stats

View full text Add to dashboard Cite

show abstract

OptiFeat: enhancing feature selection, a hybrid approach combining subject matter expertise and recursive feature elimination method

Vijayakumar,

Bharathi

2024

Discov Computing

View full text Add to dashboard Cite

Unifying mortality forecasting model: an investigation of the COM–Poisson distribution in the GAS model for improved projections

Rakhmawan,

Mahmood,

Abbas

et al. 2024

Lifetime Data Anal

View full text Add to dashboard Cite

Forecasting mortality rates is crucial for evaluating life insurance company solvency, especially amid disruptions caused by phenomena like COVID-19. The Lee–Carter model is commonly employed in mortality modelling; however, extensions that can encompass count data with diverse distributions, such as the Generalized Autoregressive Score (GAS) model utilizing the COM–Poisson distribution, exhibit potential for enhancing time-to-event forecasting accuracy. Using mortality data from 29 countries, this research evaluates various distributions and determines that the COM–Poisson model surpasses the Poisson, binomial, and negative binomial distributions in forecasting mortality rates. The one-step forecasting capability of the GAS model offers distinct advantages, while the COM–Poisson distribution demonstrates enhanced flexibility and versatility by accommodating various distributions, including Poisson and negative binomial. Ultimately, the study determines that the COM–Poisson GAS model is an effective instrument for examining time series data on mortality rates, particularly when facing time-varying parameters and non-conventional data distributions.

show abstract

Dispersion modelling of mortality for both sexes with Tweedie distributions

Cited by 4 publications

References 32 publications

Optimal Neighborhood Selection for AR-ARCH Random Fields with Application to Mortality

Optimal Neighborhood Selection for AR-ARCH Random Fields with Application to Mortality

OptiFeat: enhancing feature selection, a hybrid approach combining subject matter expertise and recursive feature elimination method

Unifying mortality forecasting model: an investigation of the COM–Poisson distribution in the GAS model for improved projections

Contact Info

Product

Resources

About