Year: 2026
Type: preprint
Abstract: Major life events have proven difficult to predict. Does this reflect limits of theory, data, and algorithms, or the large role of chance? We examine one outcome - having a child within three years - through a near-ideal setting for prediction: a data challenge where 147 researchers predicted births for Dutch residents aged 18-45, using survey data and full-population registers. Methods ranged from logistic regression to a large language model and transformers. Predictions were moderately accurate (best F1: register 0.59, survey 0.76); advanced models did not outperform classical ones; and the larger registers did not beat the survey. Simulating the stochastic biology of conception and pregnancy, we estimated a predictive ceiling (survey F1 ~ 0.86-0.94, register 0.88-0.96). Observed performance falls short of this ceiling, implicating imperfect data, methods, and unmodelled chance, while the ceiling itself shows that chance in reproduction alone sets a non-trivial limit on predicting individual lives.
Source: arXiv (Cornell University)
Authors: Sivak, Elizaveta, Cantrell, Emily M., Emery, Thomas, Garcia-Bernardo, Javier, Hafner, Flavio, Karpinska, Kasia, Lüken, Malte, Mendrik, Adrienne, Mulder, Joris, Ren, Hanzhang, Satish, Varun, Verhagen, Mark, Maineri, Angelica M., Pankowska, Paulina, Ghany, Jasmin Abdel, Arpino, Bruno, Cassani, Giovanni, Hellstrand, Julia, Ivanova, Katya, Kuikka, Sanni, Macanovic, Ana, Rahal, Charles, Tropf, Felix C., Veen, Roland J., Walasek, Nicole, van Wijk, Daniël, Wright, Kelsey Q., Zagheni, Emilio, Abbink, Henry, Aliverti, Emanuele, Amestoy, Matteo, Atav, Tilbe, Barban, Nicola, Billingsley, Sunnee, Booij, Goan J., Boucherie, Louis, Broos, Yael, Chang, Li Ya, Chiu, Jamie C., Comolli, Chiara Ludovica, Cule, Boris, Fang, Qixiang, Feehan, Dennis M., Ganly, Rachel, Gielens, Erwin, Martinez, Rolando M. Gonzales, Gradassi, Andrea, Guerra-Urzola, Rosember, Guerra-Urzola, Mario, Guerrier, Stéphane, Hassan, Enamul, Haverhoek, Vincent A., Hendrickson, Andrew T., Howard, Amber, Jin, Yuxuan, Kapoor, Sayash, van Kesteren, Erik-Jan, Klooster, Iris ten, Labussiere, Marie, Liu, Lydia T., Liu, Tiffany, Maghout, Adam, Meneghello, Simone, Mohr, Lasse, Mulder, Clara H., Newman, Saul J., Nisén, Jessica, Norden, Janis, Odgaard, Mikkel, Omenti, Riccardo, Ozdemir, Ozancan, Pao, Christina, Park, Paige, Penta, Gaia, Perdomo, Juan C., Pial, Tanzir, Piraccini, Alessio, Querin, Federica, Rao, Ziwei, Rellama, Christian, Remund, Adrien, Richert, Frederieke, van de Rijt, Arnout, Kandroodi, Mojtaba Rostami, Rotman, Stijn J., Sage, Lucas, Savcisens, Germans, Schwanitz, Katrin, Skiena, Steven, Spata, Alessandro, Stadtfeld, Yannick, Stroebl, Benedikt, Tedesco, Gaetano, Theelen, Mathilde, Tori, Gianluca, Tun-Mendicuti, Abigail, Tyagi, Rishabh, Vafa, Keyon, Vecchietti, Luiz Felipe, Vecgaile, Linda, Vermeulen, Willem R. J., Feser, Maria-Pia Victoria, Voirol, Lionel A., Volker, Thom B., Wang, Xinran, Yan, Jiani, Zhao, Xinyi, Zhou, Flora, Zilincikova, Zuzana, Nissim, Malvina, Salganik, Matthew J., Stulp, Gert
Institutions: University of Groningen, Interuniversity Center for Social Science Theory and Methodology, Princeton University, Erasmus University Rotterdam, Utrecht University, Netherlands eScience Center, Northeastern University, Tilburg University, Stanford University, University of Oxford, University of Padua, University of Helsinki, European University Institute, University College London, University of Amsterdam, Max Planck Institute for Demographic Research, Maastricht University, Vrije Universiteit Amsterdam, University of Bologna, Stockholm University, University of Geneva, Netherlands Interdisciplinary Demographic Institute, Stony Brook University, University of California, Berkeley, Technical University of Denmark, University of Turku, University of Copenhagen, Harvard University, Toulouse School of Economics, Institute for Advanced Study in Toulouse, Université Toulouse Capitole, Tallinn University, Technische Universität Dresden
Language: English
Subfield: Demography
Field: Social Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7206186786