Insurance ratemaking using a combined quantile regression machine learning 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">MAP20260012285</controlfield>
    <controlfield tag="003">MAP</controlfield>
    <controlfield tag="005">20260422174219.0</controlfield>
    <controlfield tag="008">260421e20260316esp|||p      |0|||b|spa 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=" " ind2=" ">
      <subfield code="0">MAPA20260007311</subfield>
      <subfield code="a">Zanzouri, Souleima</subfield>
    </datafield>
    <datafield tag="245" ind1="1" ind2="0">
      <subfield code="a">Insurance ratemaking using a combined quantile regression machine learning approach</subfield>
      <subfield code="c">Souleima Zanzouri, Manel Kacem and Lotfi Belkacem</subfield>
    </datafield>
    <datafield tag="520" ind1=" " ind2=" ">
      <subfield code="a">The article analyzes the process of ratemaking in non-life insurance through the combination of quantile regression and machine learning techniques. A two-stage approach is proposed that first models the probability of a claim and subsequently the aggregated claim amount, using advanced models such as random forest, gradient boosting, XGBoost, and neural networks applied to quantile regression. The empirical study, based on real automobile insurance data, shows that these models allow for a better capture of extreme risks and improve the accuracy of premium estimation. In addition, model performance is evaluated using statistical metrics, and practical implications for actuarial ratemaking are discussed</subfield>
    </datafield>
    <datafield tag="650" ind1=" " ind2="4">
      <subfield code="0">MAPA20080564322</subfield>
      <subfield code="a">Tarificación</subfield>
    </datafield>
    <datafield tag="650" ind1=" " ind2="4">
      <subfield code="0">MAPA20080573935</subfield>
      <subfield code="a">Seguros no vida</subfield>
    </datafield>
    <datafield tag="650" ind1=" " ind2="4">
      <subfield code="0">MAPA20080592011</subfield>
      <subfield code="a">Modelos actuariales</subfield>
    </datafield>
    <datafield tag="650" ind1=" " ind2="4">
      <subfield code="0">MAPA20080581886</subfield>
      <subfield code="a">Primas de seguros</subfield>
    </datafield>
    <datafield tag="650" ind1=" " ind2="4">
      <subfield code="0">MAPA20170005476</subfield>
      <subfield code="a">Machine learning</subfield>
    </datafield>
    <datafield tag="700" ind1="1" ind2=" ">
      <subfield code="0">MAPA20260007328</subfield>
      <subfield code="a">Kacem, Manel</subfield>
    </datafield>
    <datafield tag="700" ind1="1" ind2=" ">
      <subfield code="0">MAPA20260007335</subfield>
      <subfield code="a">Belkacem, Lotfi</subfield>
    </datafield>
    <datafield tag="773" ind1="0" ind2=" ">
      <subfield code="w">MAP20077000239</subfield>
      <subfield code="g">16/03/2026 Tomo 30 Número 1 - 2026 , 51 p.</subfield>
      <subfield code="x">1092-0277</subfield>
      <subfield code="t">North American actuarial journal</subfield>
      <subfield code="d">Schaumburg : Society of Actuaries, 1997-</subfield>
    </datafield>
  </record>
</collection>