Publications
Pérez Izquierdo, T. and Pronkina, E.
Economics of Transition and Institutional Change
2023
Behind the curtain: How did women's work history vary across Central and Eastern Europe?
This paper investigates the differences in female work experience across Central and Eastern European countries (CEECs). We use retrospective SHARELIFE data to analyse women's work history from 1950 to 1990. We provide descriptive evidence that women's work experience varied across CEECs. Furthermore, we argue that comparing the former provinces of the Russian Empire in Lithuania and Poland provides a natural experiment, allowing us to disentangle the effect of the differential implementation of the Soviet regime from the pre-existing differences. We find that during communism, Lithuanian women worked 2 years more by age 50 relative to their Polish counterparts. This effect is one-third of that found in the East-West Germany comparison. We propose several potential mechanisms behind this finding: the degree of land collectivization, the Church's influence and the sectoral composition. Accordingly, this study's findings highlight the importance of country differences in CEECs.
Working papers
Cuerno, M.; Galáz-García, F.; Galáz-García, S.; and Pérez Izquierdo, T.
Finding patterns of meaning: Reassessing Construal Clustering via Bipolar Class Analysis
Empirical research on construals-social affinity groups that share similar patterns of meaning--has advanced significantly in recent years. This progress is largely driven by the development of Construal Clustering Methods (CCMs), which group survey respondents into construal clusters based on similarities in their response patterns. We identify key limitations of existing CCMs, which affect their accuracy when applied to the typical structures of available data, and introduce Bipolar Class Analysis (BCA), a CCM designed to address these shortcomings. BCA measures similarity in response shifts between expressions of support and rejection across survey respondents, addressing conceptual and measurement challenges in existing methods. We formally define BCA and demonstrate its advantages through extensive simulation analyses, where it consistently outperforms existing CCMs in accurately identifying construals. Along the way, we develop a novel data-generation process that approximates more closely how individuals map latent opinions onto observable survey responses, as well as a new metric to evaluate the performance of CCMs. Additionally, we find that applying BCA to previously studied real-world datasets reveals substantively different construal patterns compared to those generated by existing CCMs in prior empirical analyses. Finally, we discuss limitations of BCA and outline directions for future research.
Escanciano, J. C. and Pérez Izquierdo, T.
Automatic Locally Robust Estimation with Generated Regressors
Many economic and causal parameters of interest depend on generated regressors. Examples include structural parameters in models with endogenous variables estimated by control functions and in models with sample selection, treatment effect estimation with propensity score matching, and marginal treatment effects. Inference with generated regressors is complicated by the very complex expression for influence functions and asymptotic variances. To address this problem, we propose Automatic Locally Robust/debiased GMM estimators in a general setting with generated regressors. Importantly, we allow for the generated regressors to be generated from machine learners, such as Random Forest, Neural Nets, Boosting, and many others. We use our results to construct novel Doubly Robust and Locally Robust estimators for the Counterfactual Average Structural Function and Average Partial Effects in models with endogeneity and sample selection, respectively. We provide sufficient conditions for the asymptotic normality of our debiased GMM estimators and investigate their finite sample performance through Monte Carlo simulations.
Pérez Izquierdo, T.; Barrio, I.; and Esteban, C.
Dynamic prediction of death risk given a renewal hospitalization process
Predicting the risk of death for chronic patients is highly valuable for informed medical decision-making. This paper proposes a general framework for dynamic prediction of the risk of death of a patient given her hospitalization history, which is generally available to physicians. Predictions are based on a joint model for the death and hospitalization processes, thereby avoiding the potential bias arising from selection of survivors. The framework accommodates various submodels for the hospitalization process. In particular, we study prediction of the risk of death in a renewal model for hospitalizations, a common approach to recurrent event modelling. In the renewal model, the distribution of hospitalizations throughout the follow-up period impacts the risk of death. This result differs from prediction in the Poisson model, previously studied, where only the number of hospitalizations matters. We apply our methodology to a prospective, observational cohort study of 512 patients treated for COPD in one of six outpatient respiratory clinics run by the Respiratory Service of Galdakao University Hospital, with a median follow-up of 4.7 years. We find that more concentrated hospitalizations increase the risk of death.
Pérez Izquierdo, T.
The determinants of counterfactual identification in the binary choice model with endogenous regressors
The Counterfactual Average Structural Function (CASF) is the Average Structural Function (ASF) averaged with respect to a counterfactual distribution of covariates. In the binary choice model, the CASF measures the rate of successes in a counterfactual scenario. This paper shows that the CASF is non-parametrically identified as a weighted average, where the weights are given by a likelihood ratio. Estimation of the CASF at the regular root n-rate requires a square integrability condition for those weights. This necessary condition depends on instrument strength, the degree of endogeneity, and the relevance of the regressors. Much insight is gained from the single normal regressor model. In this setting, I show that extrapolation strength (the capacity of the model to regularly identify the CASF) increases with the relevance of the regressor and, surprisingly, with the degree of endogeneity. The impact of instrument strength is found to be non-monotone. Moreover, I find a threshold for instrument strength below which regular identification of the CASF is not possible.