Maha Ouali*, Simon Rebeyrolles**

*Accurately estimating treatment effects in time series is essential for evaluating interventions in real-world applications, especially when treatment assignment is biased by unobserved factors. In many practical settings, interventions are adopted at different times across individuals, leading to staggered treatment exposure and heterogeneous pre-treatment histories. In such cases, aggregating outcome trajectories across treated units is ill-defined, making individual treatment effect (ITE) estimation a prerequisite for reliable causal inference. We therefore study the problem of estimating the average treatment effect for the treated (ATT) by first recovering individual-level counterfactuals. We introduce a neural framework that learns simultaneously low-dimensional latent representations of individual time series and propensity scores. These estimates are then used to approximate the individual treatment effects through a flexible matching procedure that avoids classical convexity constraints commonly used in synthetic control methods. By operating at the individual level, our approach naturally accommodates staggered interventions and improves counterfactual estimation under latent bias, without relying on explicit temporal modeling assumptions. We illustrate our approach on both real-world energy consumption data and clinical time series, including high-frequency electricity demand-response programs and semi-synthetic data for individuals in intensive care unit (ICU), where hidden confounding, staggered treatment adoption, and non-stationary dynamics are prevalent.

**This paper provides new evidence on the effects of access to public housing in France on a large set of variables, such as housing quality, neighborhood characteristics, and labor outcomes. To this end, a unique and novel linkage between data on social housing applicants and tax records is implemented. The empirical strategy exploits the fact that, in areas where the imbalance between the supply and demand for social housing is substantial, allocations can occur randomly among applicants with similar observable characteristics. To approximate this quasi-experimental setting, I combine propensity score matching between recipients and non-recipients with a staggered difference-in-differences design. The results suggest that recipients of social housing experience a trade-off in their living conditions. In particular, their housing quality improves substantially, notably through reductions in overcrowding and rent burdens. However, the likelihood of residing in a disadvantaged neighborhood increases following allocation. Although the theoretical literature predicts deteriorated labor market outcomes due to a disincentive channel or spatial mismatch, no significant impact is observed on recipients’ employment trajectories in the short run, regardless of the dimensions of heterogeneity considered.