What to Forecast? An Exploratory Study of the Implication of Using Different Indicators to Forecast Mortality

Publikation: Konferencebidrag uden forlag/tidsskriftKonferenceabstrakt til konferenceForskningpeer review

Resumé

Researchers and private and public institutions have suggested different ways to forecast mortality over the years. Many of these forecasting models are based on extrapolative methods of the age-specific death rates. However, more recent studies have looked into forecasting models based on other mortality indicators, such as life expectancy or the life table deaths. We here ask what are the implication of choosing one specific indicator over another to forecast mortality? We compare five models based on singular value decomposition of the matrix by age and time of different indicators: logged death rates, logit of death probabilities, logit of survival probabilities, log-ratio transformation of the life table deaths and life expectancy at birth. The results show that the time-indexes of all these indicators are very similar, but that forecasting using death rates and probabilities of death leads to more pessimistic forecasts than the other models.
OriginalsprogEngelsk
Publikationsdato1. apr. 2017
StatusUdgivet - 1. apr. 2017
BegivenhedAnnual Meeting of the Population Association of America 2017 - Chicago, USA
Varighed: 27. apr. 201729. apr. 2017

Konference

KonferenceAnnual Meeting of the Population Association of America 2017
LandUSA
ByChicago
Periode27/04/201729/04/2017

Fingeraftryk

mortality
life table
life expectancy
decomposition
matrix
forecast
indicator
rate
public institution
index
method

Citer dette

@conference{da4aa58cbe31490eb550b3869f02d897,
title = "What to Forecast? An Exploratory Study of the Implication of Using Different Indicators to Forecast Mortality",
abstract = "Researchers and private and public institutions have suggested different ways to forecast mortality over the years. Many of these forecasting models are based on extrapolative methods of the age-specific death rates. However, more recent studies have looked into forecasting models based on other mortality indicators, such as life expectancy or the life table deaths. We here ask what are the implication of choosing one specific indicator over another to forecast mortality? We compare five models based on singular value decomposition of the matrix by age and time of different indicators: logged death rates, logit of death probabilities, logit of survival probabilities, log-ratio transformation of the life table deaths and life expectancy at birth. The results show that the time-indexes of all these indicators are very similar, but that forecasting using death rates and probabilities of death leads to more pessimistic forecasts than the other models.",
author = "{Bergeron Boucher}, Marie-Pier and S{\o}ren Kj{\ae}rgaard and James Oeppen",
year = "2017",
month = "4",
day = "1",
language = "English",
note = "Annual Meeting of the Population Association of America 2017 ; Conference date: 27-04-2017 Through 29-04-2017",

}

Bergeron Boucher, M-P, Kjærgaard, S & Oeppen, J 2017, 'What to Forecast? An Exploratory Study of the Implication of Using Different Indicators to Forecast Mortality', Annual Meeting of the Population Association of America 2017, Chicago, USA, 27/04/2017 - 29/04/2017.

What to Forecast? An Exploratory Study of the Implication of Using Different Indicators to Forecast Mortality. / Bergeron Boucher, Marie-Pier; Kjærgaard, Søren; Oeppen, James.

2017. Abstract fra Annual Meeting of the Population Association of America 2017, Chicago, USA.

Publikation: Konferencebidrag uden forlag/tidsskriftKonferenceabstrakt til konferenceForskningpeer review

TY - ABST

T1 - What to Forecast? An Exploratory Study of the Implication of Using Different Indicators to Forecast Mortality

AU - Bergeron Boucher, Marie-Pier

AU - Kjærgaard, Søren

AU - Oeppen, James

PY - 2017/4/1

Y1 - 2017/4/1

N2 - Researchers and private and public institutions have suggested different ways to forecast mortality over the years. Many of these forecasting models are based on extrapolative methods of the age-specific death rates. However, more recent studies have looked into forecasting models based on other mortality indicators, such as life expectancy or the life table deaths. We here ask what are the implication of choosing one specific indicator over another to forecast mortality? We compare five models based on singular value decomposition of the matrix by age and time of different indicators: logged death rates, logit of death probabilities, logit of survival probabilities, log-ratio transformation of the life table deaths and life expectancy at birth. The results show that the time-indexes of all these indicators are very similar, but that forecasting using death rates and probabilities of death leads to more pessimistic forecasts than the other models.

AB - Researchers and private and public institutions have suggested different ways to forecast mortality over the years. Many of these forecasting models are based on extrapolative methods of the age-specific death rates. However, more recent studies have looked into forecasting models based on other mortality indicators, such as life expectancy or the life table deaths. We here ask what are the implication of choosing one specific indicator over another to forecast mortality? We compare five models based on singular value decomposition of the matrix by age and time of different indicators: logged death rates, logit of death probabilities, logit of survival probabilities, log-ratio transformation of the life table deaths and life expectancy at birth. The results show that the time-indexes of all these indicators are very similar, but that forecasting using death rates and probabilities of death leads to more pessimistic forecasts than the other models.

M3 - Conference abstract for conference

ER -