Gaussian Process Models for Mortality Rates and Improvement Factors

Ludkovski, Michael; Risk, James; Zail, Howard

doi:10.2139/ssrn.2831831

Cited by 8 publications

(11 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These remarks are coherent with the classical approach to mortality modeling and forecasting, in which mortality rates are often represented by an age dependent function, whose evolution over time is described using time-series (see e.g. the short model review in Ludkovski et al (2016)). However, if the period dimension appears to be well-suited to capture mortality changes such as a cause-of-death mortality reduction, we show in the following that variations in mortality caused by changes in population composition are not necessarily well-captured by period indexes.…”

Section: Cause Of Death Reduction and Compensation Effectsupporting

confidence: 53%

How can a cause-of-death reduction be compensated for by the population heterogeneity? A dynamic approach

Kaakaï

Hardy²,

Arnold

2019

Insurance: Mathematics and Economics

View full text Add to dashboard Cite

A growing number of studies indicate a widening of socioeconomic inequalities in mortality over the past decades. It has therefore become crucially important to understand the impact of heterogeneity and its evolution on the future mortality of heterogeneous populations. In particular, recent developments in multi-population mortality have raised a number of questions, among which is the issue of evaluating cause-of-death reduction targets set by national and international institutions in the presence of heterogeneity. The aim of this paper is to show how the study of the population data and the population dynamics framework contribute to addressing these issues, by providing a new viewpoint on the evolution of aggregate mortality indicators in the presence of heterogeneity. Our findings rely on two unique datasets on the English population and cause-specific number of deaths by socioeconomic circumstances, over the period 1981-2015. The analysis of the data first highlights the complexity of recent demographic developments, characterized by significant composition changes in the population, with considerable variations according to the age class or cohort, along with a widening of socioeconomic inequalities. We then introduce a dynamic framework for studying the impact of composition changes on the mortality of the global population. In particular, we are interested in quantifying the impacts of cause-of-death mortality reduction in comparison with changes of composition in a heterogeneous population. We show how a cause of death reduction could be compensated for in the presence of heterogeneity, which could lead to misinterpretations when assessing public policies impacts and/or for the forecasting of future trends.

show abstract

Section: Cause Of Death Reduction and Compensation Effectsupporting

confidence: 53%

How can a cause-of-death reduction be compensated for by the population heterogeneity? A dynamic approach

Kaakaï

Hardy²,

Arnold

2019

Insurance: Mathematics and Economics

View full text Add to dashboard Cite

show abstract

“…Although Gaussian processes themselves have long been an integral part of time-series econometrics, applications of Gaussian-process regression have only recently appeared in the econometrics literature. For example, Kasy (2018) uses Gaussian-process regression to model the relationship between healthcare expenditures and coinsurance levels and Ludkovski et al (2018) to model mortality rates as a function of the individual's age and calendar year. Ruseckaite et al (2018) use multivariate Gaussian-process regression in a mixture-amount model to capture the dependence of the mixture parameters on the amounts themselves.…”

Section: Introductionmentioning

confidence: 99%

Semi-parametric analysis of efficiency and productivity using Gaussian processes

Emvalomatis

2019

The Econometrics Journal

View full text Add to dashboard Cite

Summary This paper proposes a fully Bayesian semi-parametric method for efficiency and productivity analysis based on Gaussian processes. The proposed technique frees the researcher from having to specify a functional form for the production frontier, and it is shown in simulated data to perform as well as flexible parametric models when correct distributional assumptions are imposed on the inefficiency component of the error term, and slightly better when incorrect assumptions are made. The technique is applied to a panel dataset of US electric utilities, where total-factor productivity growth is estimated and decomposed with both parametric and semi-parametric techniques.

show abstract

“…Coregionalization is a dimension reduction technique that enables efficiently handling many correlated outputs. This work is a continuation of our series of articles Ludkovski et al (2018), Huynh et al (2020) and Huynh and Ludkovski (2021) that discussed the application of GPs to model all-cause mortality in the single-population and multi-population contexts, respectively. Unlike all-cause mortality in different geographic regions, which tends to exhibit strong correlation and long-term coherence, different causes have less commonality, and thus require a more flexible structure for the respective cross-dependence.…”

Section: Background and Motivationmentioning

confidence: 88%

Joint Models for Cause-of-Death Mortality in Multiple Populations

Huynh¹,

Ludkovski²

2021

Preprint

View full text Add to dashboard Cite

We investigate jointly modeling Age-specific rates of various causes of death in a multinational setting. We apply Multi-Output Gaussian Processes (MOGP), a spatial machine learning method, to smooth and extrapolate multiple cause-of-death mortality rates across several countries and both genders. To maintain flexibility and scalability, we investigate MOGPs with Kronecker-structured kernels and latent factors. In particular, we develop a custom multi-level MOGP that leverages the gridded structure of mortality tables to efficiently capture heterogeneity and dependence across different factor inputs. Results are illustrated with datasets from the Human Cause-of-Death Database (HCD). We discuss a case study involving cancer variations in three European nations, and a US-based study that considers eight top-level causes and includes comparison to all-cause analysis. Our models provide insights into the commonality of cause-specific mortality trends and demonstrate the opportunities for respective data fusion.

show abstract

Gaussian Process Models for Mortality Rates and Improvement Factors

Cited by 8 publications

References 24 publications

How can a cause-of-death reduction be compensated for by the population heterogeneity? A dynamic approach

How can a cause-of-death reduction be compensated for by the population heterogeneity? A dynamic approach

Semi-parametric analysis of efficiency and productivity using Gaussian processes

Joint Models for Cause-of-Death Mortality in Multiple Populations

Contact Info

Product

Resources

About