Multi-population mortality forecasting using tensor decomposition

Dong, Yumo; Huang, Fei; Yu, Honglin; Haberman, Steven

doi:10.1080/03461238.2020.1740314

Cited by 27 publications

(15 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…., and 10 one-step-ahead forecasts. We compare the proposed method with the high-dimensional functional time series (HDFTS, Gao et al, 2019), the high-dimensional functional factor model (HDFFM, Nisol et al, 2019), the factor models for matrix-valued high-dimensional time series (MFM, Wang et al, 2019) and the tensor decomposition forecast methods (Canonical Polyadic Decomposition (CPD) and Tucker) of Dong et al (2020).…”

Section: Forecast Evaluationmentioning

confidence: 99%

“…In the empirical analysis, we compare the in sample forecasting performance of the proposed model with that of the aforementioned functional factor models and several other multi-population mortality models (Dong et al, 2020). We find that the proposed high-dimensional functional factor model produces more accurate point and interval forecasts than the other models.…”

Section: Introductionmentioning

confidence: 98%

See 1 more Smart Citation

Multi-population mortality forecasting using high-dimensional functional factor models

Tang¹,

Shang²,

Yang³

2021

Preprint

View full text Add to dashboard Cite

This paper proposes a two-fold factor model for high-dimensional functional time series (HDFTS), which enables the modeling and forecasting of multi-population mortality under the functional data framework. The proposed model first decomposes the HDFTS into functional time series with lower dimensions (common feature) and a system of basis functions specific to different cross-sections (heterogeneity). Then the lower-dimensional common functional time series are further reduced into low-dimensional scalar factor matrices. The dimensionally reduced factor matrices can reasonably convey useful information in the original HDFTS. All the temporal dynamics contained in the original HDFTS are extracted to facilitate forecasting. The proposed model can be regarded as a general case of several existing functional factor models. Through a Monte Carlo simulation, we demonstrate the performance of the proposed method in model fitting. In an empirical study of the Japanese subnational age-specific mortality rates, we show that the proposed model produces more accurate point and interval forecasts in modeling multi-population mortality than those existing functional factor models. The financial impact of the improvements in forecasts is demonstrated through comparisons in life annuity pricing practices.

show abstract

Section: Forecast Evaluationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 98%

Multi-population mortality forecasting using high-dimensional functional factor models

Tang¹,

Shang²,

Yang³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…While there have been many works on modeling mortality across several populations (Dong et al 2020, Enchev et al 2017, Guibert et al 2019, Hyndman et al 2013, Kleinow 2015, Li and Lu 2017, Tsai and Zhang 2019, as well as an active literature on cause-of-death mortality, there are very few that do both simultaneously. As we detail below, there are many natural reasons for building such a joint model, and this gap is arguably driven by the underlying "Big Data" methodological challenge.…”

Section: Background and Motivationmentioning

confidence: 99%

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

“…Even though Debón et al [9] used some spatial techniques on residuals to forecast dynamic life tables, the importance of the spatial effect has not been well recognized in modeling and forecasting dynamics of mortality. In recent years, there has been research on multi-population mortality models, for example, using a group of countries with similar social-economic status, or males and females in the same population [10]. However, these models do not measure spatial dependence and temporal and spatial effects obtained with data panel models.…”

Section: Introductionmentioning

confidence: 99%

Spatiotemporal Econometrics Models for Old Age Mortality in Europe

Carracedo

Debón

2021

Mathematics

View full text Add to dashboard Cite

In the past decade, panel data models using time-series observations of several geographical units have become popular due to the availability of software able to implement them. The aim of this study is an updated comparison of estimation techniques between the implementations of spatiotemporal panel data models across MATLAB and R softwares in order to fit real mortality data. The case study used concerns the male and female mortality of the aged population of European countries. Mortality is quantified with the Comparative Mortality Figure, which is the most suitable statistic for comparing mortality by sex over space when detailed specific mortality is available for each studied population. The spatial dependence between the 26 European countries and their neighbors during 1995–2012 was confirmed through the Global Moran Index and the spatiotemporal panel data models. For this reason, it can be said that mortality in European population aging not only depends on differences in the health systems, which are subject to national discretion but also on supra-national developments. Finally, we conclude that although both programs seem similar, there are some differences in the estimation of parameters and goodness of fit measures being more reliable MATLAB. These differences have been justified by detailing the advantages and disadvantages of using each of them.

show abstract

Multi-population mortality forecasting using tensor decomposition

Cited by 27 publications

References 32 publications

Multi-population mortality forecasting using high-dimensional functional factor models

Multi-population mortality forecasting using high-dimensional functional factor models

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

Spatiotemporal Econometrics Models for Old Age Mortality in Europe

Contact Info

Product

Resources

About