IDEAS home Printed from https://ideas.repec.org/p/arx/papers/2506.23619.html
   My bibliography  Save this paper

Overparametrized models with posterior drift

Author

Listed:
  • Guillaume Coqueret
  • Martial Laguerre

Abstract

This paper investigates the impact of posterior drift on out-of-sample forecasting accuracy in overparametrized machine learning models. We document the loss in performance when the loadings of the data generating process change between the training and testing samples. This matters crucially in settings in which regime changes are likely to occur, for instance, in financial markets. Applied to equity premium forecasting, our results underline the sensitivity of a market timing strategy to sub-periods and to the bandwidth parameters that control the complexity of the model. For the average investor, we find that focusing on holding periods of 15 years can generate very heterogeneous returns, especially for small bandwidths. Large bandwidths yield much more consistent outcomes, but are far less appealing from a risk-adjusted return standpoint. All in all, our findings tend to recommend cautiousness when resorting to large linear models for stock market predictions.

Suggested Citation

  • Guillaume Coqueret & Martial Laguerre, 2025. "Overparametrized models with posterior drift," Papers 2506.23619, arXiv.org.
  • Handle: RePEc:arx:papers:2506.23619
    as

    Download full text from publisher

    File URL: http://arxiv.org/pdf/2506.23619
    File Function: Latest version
    Download Restriction: no
    ---><---

    References listed on IDEAS

    as
    1. Ivo Welch & Amit Goyal, 2008. "A Comprehensive Look at The Empirical Performance of Equity Premium Prediction," The Review of Financial Studies, Society for Financial Studies, vol. 21(4), pages 1455-1508, July.
    2. Patrick Gagliardini & Elisa Ossola & Olivier Scaillet, 2016. "Time‐Varying Risk Premium in Large Cross‐Sectional Equity Data Sets," Econometrica, Econometric Society, vol. 84, pages 985-1046, May.
    3. Klaus Frick & Axel Munk & Hannes Sieling, 2014. "Multiscale change point inference," Journal of the Royal Statistical Society Series B, Royal Statistical Society, vol. 76(3), pages 495-580, June.
    4. Bakalli, Gaetan & Guerrier, Stéphane & Scaillet, Olivier, 2023. "A penalized two-pass regression to predict stock returns with time-varying risk premia," Journal of Econometrics, Elsevier, vol. 237(2).
    5. repec:bla:jfinan:v:53:y:1998:i:2:p:549-573 is not listed on IDEAS
    6. Dangl, Thomas & Halling, Michael, 2012. "Predictive regressions with time-varying coefficients," Journal of Financial Economics, Elsevier, vol. 106(1), pages 157-181.
    7. Leland E. Farmer & Lawrence Schmidt & Allan Timmermann, 2023. "Pockets of Predictability," Journal of Finance, American Finance Association, vol. 78(3), pages 1279-1341, June.
    8. Stefan Nagel, 2025. "Seemingly Virtuous Complexity in Return Prediction," NBER Working Papers 34104, National Bureau of Economic Research, Inc.
    9. Silverstein, J. W., 1995. "Strong Convergence of the Empirical Distribution of Eigenvalues of Large Dimensional Random Matrices," Journal of Multivariate Analysis, Elsevier, vol. 55(2), pages 331-339, November.
    10. Subha Maity & Diptavo Dutta & Jonathan Terhorst & Yuekai Sun & Moulinath Banerjee, 2024. "A linear adjustment-based approach to posterior drift in transfer learning," Biometrika, Biometrika Trust, vol. 111(1), pages 31-50.
    11. Tim Bollerslev & Jia Li & Yuexuan Ren, 2024. "Optimal Inference for Spot Regressions," American Economic Review, American Economic Association, vol. 114(3), pages 678-708, March.
    12. Bryan Kelly & Semyon Malamud & Kangying Zhou, 2024. "The Virtue of Complexity in Return Prediction," Journal of Finance, American Finance Association, vol. 79(1), pages 459-503, February.
    Full references (including those not matched with items on IDEAS)

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Jinbo Cai & Wenze Li & Wenjie Wang, 2025. "Electricity Market Predictability: Virtues of Machine Learning and Links to the Macroeconomy," Papers 2507.07477, arXiv.org.
    2. Smith, Simon C., 2021. "International stock return predictability," International Review of Financial Analysis, Elsevier, vol. 78(C).
    3. Demetrescu, Matei & Georgiev, Iliyan & Rodrigues, Paulo M.M. & Taylor, A.M. Robert, 2022. "Testing for episodic predictability in stock returns," Journal of Econometrics, Elsevier, vol. 227(1), pages 85-113.
    4. Lof, Matthijs & Nyberg, Henri, 2024. "Discount rates and cash flows: A local projection approach," Journal of Banking & Finance, Elsevier, vol. 162(C).
    5. Timmermann, Allan, 2018. "Forecasting Methods in Finance," CEPR Discussion Papers 12692, C.E.P.R. Discussion Papers.
    6. Guillaume Coqueret, 2023. "Forking paths in financial economics," Papers 2401.08606, arXiv.org.
    7. Faria, Gonçalo & Verona, Fabio, 2024. "Enhancing forecast accuracy through frequencydomain combination: Applications to financial and economic indicators," Bank of Finland Research Discussion Papers 14/2024, Bank of Finland.
    8. Schlosky, Minh Tam Tammy & Karadas, Serkan & Stivers, Adam, 2024. "Forecasting U.S. Stock Returns Conditional on Geopolitical Risk and Business Cycles," International Review of Financial Analysis, Elsevier, vol. 96(PB).
    9. Cheng, Tingting & Jiang, Shan & Zhao, Albert Bo & Zhao, Junyi, 2025. "Is machine learning a necessity? A regression-based approach for stock return prediction," Journal of Empirical Finance, Elsevier, vol. 81(C).
    10. Allan Timmermann, 2018. "Forecasting Methods in Finance," Annual Review of Financial Economics, Annual Reviews, vol. 10(1), pages 449-479, November.
    11. Feng, Guanhao & He, Xin & Wang, Yanchu & Wu, Chunchi, 2025. "Predicting individual corporate bond returns," Journal of Banking & Finance, Elsevier, vol. 171(C).
    12. Davide Pettenuzzo & Francesco Ravazzolo, 2016. "Optimal Portfolio Choice Under Decision‐Based Model Combinations," Journal of Applied Econometrics, John Wiley & Sons, Ltd., vol. 31(7), pages 1312-1332, November.
    13. Yi, Yongsheng & Ma, Feng & Zhang, Yaojie & Huang, Dengshi, 2019. "Forecasting stock returns with cycle-decomposed predictors," International Review of Financial Analysis, Elsevier, vol. 64(C), pages 250-261.
    14. Daniel de Almeida & Ana-Maria Fuertes & Luiz Koodi Hotta, 2025. "Out-of-Sample Predictability of the Equity Risk Premium," Mathematics, MDPI, vol. 13(2), pages 1-23, January.
    15. Haibin Xie & Shouyang Wang, 2015. "Risk-return trade-off, information diffusion, and U.S. stock market predictability," International Journal of Financial Engineering (IJFE), World Scientific Publishing Co. Pte. Ltd., vol. 2(04), pages 1-20, December.
    16. Goodness C. Aye & Mehmet Balcilar & Rangan Gupta, 2017. "International stock return predictability: Is the role of U.S. time-varying?," Empirica, Springer;Austrian Institute for Economic Research;Austrian Economic Association, vol. 44(1), pages 121-146, February.
    17. Chen, Yong & Da, Zhi & Huang, Dayong, 2022. "Short selling efficiency," Journal of Financial Economics, Elsevier, vol. 145(2), pages 387-408.
    18. José Afonso Faias & Juan Arismendi Zambrano, 2022. "Equity Risk Premium Predictability from Cross-Sectoral Downturns [International asset allocation with regime shifts]," The Review of Asset Pricing Studies, Society for Financial Studies, vol. 12(3), pages 808-842.
    19. repec:ipg:wpaper:2013-020 is not listed on IDEAS
    20. Nonejad, Nima, 2022. "Predicting equity premium out-of-sample by conditioning on newspaper-based uncertainty measures: A comparative study," International Review of Financial Analysis, Elsevier, vol. 83(C).
    21. Liu, Li & Ma, Feng & Wang, Yudong, 2015. "Forecasting excess stock returns with crude oil market data," Energy Economics, Elsevier, vol. 48(C), pages 316-324.

    More about this item

    NEP fields

    This paper has been announced in the following NEP Reports:

    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:arx:papers:2506.23619. 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.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with 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: arXiv administrators (email available below). General contact details of provider: http://arxiv.org/ .

    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.