Extending the lee-carter model with variational autoencoder: a fusion of neural network and bayesian approach

<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.loc.gov/MARC21/slim http://www.loc.gov/standards/marcxml/schema/MARC21slim.xsd">
  <record>
    <leader>00000cab a2200000   4500</leader>
    <controlfield tag="001">MAP20220026123</controlfield>
    <controlfield tag="003">MAP</controlfield>
    <controlfield tag="005">20221004104208.0</controlfield>
    <controlfield tag="008">221004e20220905bel|||p      |0|||b|eng d</controlfield>
    <datafield tag="040" ind1=" " ind2=" ">
      <subfield code="a">MAP</subfield>
      <subfield code="b">spa</subfield>
      <subfield code="d">MAP</subfield>
    </datafield>
    <datafield tag="084" ind1=" " ind2=" ">
      <subfield code="a">6</subfield>
    </datafield>
    <datafield tag="100" ind1="1" ind2=" ">
      <subfield code="0">MAPA20220008563</subfield>
      <subfield code="a">Miyata, Akihiro</subfield>
    </datafield>
    <datafield tag="245" ind1="1" ind2="0">
      <subfield code="a">Extending the lee-carter model with variational autoencoder: a fusion of neural network and bayesian approach</subfield>
      <subfield code="c">Akihiro Miyata</subfield>
    </datafield>
    <datafield tag="520" ind1=" " ind2=" ">
      <subfield code="a">In this study, we propose a nonlinear Bayesian extension of the LeeCarter (LC) model using a single-stage procedure with a dimensionality reduction neural network (NN). LC is originally estimated using a two-stage procedure: dimensionality reduction of data by singular value decomposition followed by a time series model fitting. To address the limitations of LC, which are attributed to the two-stage estimation and insufficient model fitness to data, single-stage procedures using the Bayesian state-space (BSS) approaches and extensions of flexibility in modeling by NNs have been proposed. As a fusion of these two approaches, we propose a NN extension of LC with a variational autoencoder that performs the variational Bayesian estimation of a state-space model and dimensionality reduction by autoencoding. Despite being a NN model that performs single-stage estimation of parameters, our model has excellent interpretability and the ability to forecast with confidence intervals, as with the BSS models, without using Markov chain Monte Carlo methods.

</subfield>
    </datafield>
    <datafield tag="540" ind1=" " ind2=" ">
      <subfield code="a">La copia digital se distribuye bajo licencia "Attribution 4.0 International (CC BY 4.0)"</subfield>
      <subfield code="f"/>
      <subfield code="u">https://creativecommons.org/licenses/by/4.0</subfield>
      <subfield code="9">43</subfield>
    </datafield>
    <datafield tag="650" ind1=" " ind2="4">
      <subfield code="0">MAPA20100065273</subfield>
      <subfield code="a">Modelo Lee-Carter</subfield>
    </datafield>
    <datafield tag="650" ind1=" " ind2="4">
      <subfield code="0">MAPA20100065242</subfield>
      <subfield code="a">Teorema de Bayes</subfield>
    </datafield>
    <datafield tag="650" ind1=" " ind2="4">
      <subfield code="0">MAPA20080579258</subfield>
      <subfield code="a">Cálculo actuarial</subfield>
    </datafield>
    <datafield tag="773" ind1="0" ind2=" ">
      <subfield code="w">MAP20077000420</subfield>
      <subfield code="g">05/09/2022 Volumen 52 Número 3 - septiembre 2022 , p. 789-812</subfield>
      <subfield code="x">0515-0361</subfield>
      <subfield code="t">Astin bulletin</subfield>
      <subfield code="d">Belgium : ASTIN and AFIR Sections of the International Actuarial Association</subfield>
    </datafield>
    <datafield tag="856" ind1=" " ind2=" ">
      <subfield code="q">application/pdf</subfield>
      <subfield code="w">1116870</subfield>
      <subfield code="y">Recurso electrónico / Electronic resource</subfield>
    </datafield>
  </record>
</collection>