IDEAS home Printed from https://ideas.repec.org/a/rom/conase/v6y2024i1p631-647.html
   My bibliography  Save this article

Supervised Learning Algorithms for Non-Life SCR Ratio Forecasting

Author

Listed:
  • Marius ACATRINEI

    (Institute for Economic Forecasting, Romanian Academy, Bucharest, Romania)

  • Adriana AnaMaria DAVIDESCU

    (Bucharest University of Economic Studies, Bucharest, Romania)

  • Laurentiu Paul BARANGA

    (Bucharest University of Economic Studies, Bucharest, Romania)

  • Razvan Gabriel HAPAU

    (West University of Timisoara, Timisoara, Romania)

  • George CALIN

    (Bucharest University of Economic Studies, Bucharest, Romania)

Abstract

The solvency is measured by the Solvency Capital Requirement (SCR). This study seeks to determine the best financial ratios to forecast SCR because it is significant. There is seasonality, data jumps, and shifts in insurance indicators, which make prediction of SCR difficult. Different machine learning algorithms are applied to the insurance market in this research to see how well they can describe and predict the SCR ratio. Gaussian process regression, ensemble methods, regression decision trees, stepwise regression, and neural networks were used as supervised learning techniques to find the most suitable method to predict SCR. According to our analysis of nonlife insurance data from Romania between 2016-2020, debt ratio, reserve adequacy, receivables, and liquidity are among the key indicators that should be considered when forecasting SCR. These findings can be useful for policymakers, regulators, actuaries, and professionals involved in risk management or the insurance industry.

Suggested Citation

  • Marius ACATRINEI & Adriana AnaMaria DAVIDESCU & Laurentiu Paul BARANGA & Razvan Gabriel HAPAU & George CALIN, 2024. "Supervised Learning Algorithms for Non-Life SCR Ratio Forecasting," PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON ECONOMICS AND SOCIAL SCIENCES, Bucharest University of Economic Studies, Romania, vol. 6(1), pages 631-647, August.
    • Handle: RePEc:rom:conase:v:6:y:2024:i:1:p:631-647
    as

    Download full text from publisher

    File URL: https://www.icess.ase.ro/supervised-learning-algorithms-for-non-life-scr-ratio-forecasting/
    Download Restriction: no
    ---><---

    More about this item

    Keywords

    general insurance; machine learning; risk prediction; solvability capital requirement ratio.;
    All these keywords.

    JEL classification:

    • G22 - Financial Economics - - Financial Institutions and Services - - - Insurance; Insurance Companies; Actuarial Studies
    • G28 - Financial Economics - - Financial Institutions and Services - - - Government Policy and Regulation

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:rom:conase:v:6:y:2024:i:1:p:631-647. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    We have no bibliographic references for this item. You can help adding them by using this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Zamfir Andreea (email available below). General contact details of provider: https://edirc.repec.org/data/aseeero.html .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.