Using Text-Based Life Trajectories from Swedish Register Data to Predict Residential Mobility with Pretrained Transformers

TL;DR

This study converts Swedish register data into text sequences to predict residential mobility, demonstrating that transformer-based NLP models effectively capture individual life trajectories and improve longitudinal predictions.

Contribution

It introduces a novel method of transforming large-scale register data into textual sequences and evaluates multiple transformer models for predicting residential mobility.

Findings

Transformer models outperform baseline methods in prediction accuracy.

Textualized register data retains meaningful individual trajectory information.

Sequential models effectively capture temporal and semantic structures.

Abstract

We transform large-scale Swedish register data into textual life trajectories to address two long-standing challenges in data analysis: high cardinality of categorical variables and inconsistencies in coding schemes over time. Leveraging this uniquely comprehensive population register, we convert register data from 6.9 million individuals (2001-2013) into semantically rich texts and predict individuals' residential mobility in later years (2013-2017). These life trajectories combine demographic information with annual changes in residence, work, education, income, and family circumstances, allowing us to assess how effectively such sequences support longitudinal prediction. We compare multiple NLP architectures (including LSTM, DistilBERT, BERT, and Qwen) and find that sequential and transformer-based models capture temporal and semantic structure more effectively than baseline models. The results show that textualized register data preserves meaningful information about individual pathways and supports complex, scalable modeling. Because few countries maintain longitudinal microdata with comparable coverage and precision, this dataset enables analyses and methodological tests that would be difficult or impossible elsewhere, offering a rigorous testbed for developing and evaluating new sequence-modeling approaches. Overall, our findings demonstrate that combining semantically rich register data with modern language models can substantially advance longitudinal analysis in social sciences.