OBOtilia Boldea

Otilia Boldea

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Otilia Boldea has a total of 21 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Retrospective evaluation of school-related measures on pre-vaccination transmission dynamics of SARS-CoV-2

    by Benedetta Canfora, Rey Audie Escosio, Christiaan H. van Dorp, Otilia Boldea, Marc Bonten, Ana Nunes, Ganna Rozhnova

  • Open Access

    Counterfactual evaluation of elementary and secondary school policies in the COVID-19 pandemic

    by Benedetta Canfora, Rey Audie Escosio, Otilia Boldea, Ana Nunes, Ganna Rozhnova

    Understanding the role of elementary and secondary schools in SARS-CoV-2 transmission is essential for designing effective public health policies. We developed a stochastic age-stratified transmission model, fitted to epidemiological and contact survey data, to evaluate how alternative school closure and reopening strategies might have shaped the course of the pandemic in the Netherlands from early 2020 through late 2021. Using counterfactual simulations and time-varying elasticity analysis, we found that school closures could mitigate transmission and reduce healthcare burden. However, their effectiveness was highly context-dependent, influenced by school type, contact patterns, policy timing, and population immunity. Age groups driving transmission shifted over time, with adults dominating early in the pandemic, adolescents in late 2020 and early 2021, and children in late 2021. Accordingly, secondary schools had a greater impact on national hospitalizations in 2020, while elementary schools became more important in 2021, partly due to lower infection-induced and negligible vaccine-induced immunity among children compared to adolescents. Together, these results underscore the importance of accounting for age-specific transmission dynamics and the evolving epidemiological landscape when evaluating school closure and reopening strategies. Our study supports a more targeted and adaptive approach to school policies in future pandemics.

  • Open Access

    Consistency of Selection Strategies for Fraud Detection

    by Christos Revelas, Otilia Boldea, Bas J. M. Werker

    This paper studies how insurers can chose which claims to investigate for fraud. Given a prediction model, typically only claims with the highest predicted propability of being fraudulent are investigated. We argue that this can lead to inconsistent learning and propose a randomized alternative. More generally, we draw a parallel with the multi-arm bandit literature and argue that, in the presence of selection, the obtained observations are not iid. Hence, dependence on past observations should be accounted for when updating parameter estimates. We formalize selection in a binary regression framework and show that model updating and maximum-likelihood estimation can be implemented as if claims were investigated at random. Then, we define consistency of selection strategies and conjecture sufficient conditions for consistency. Our simulations suggest that the often-used selection strategy can be inconsistent while the proposed randomized alternative is consistent. Finally, we compare our randomized selection strategy with Thompson sampling, a standard multi-arm bandit heuristic. Our simulations suggest that the latter can be inefficient in learning low fraud probabilities.

  • Open Access

    Counterfactual evaluation of elementary and secondary school policies in the COVID-19 pandemic

    by Benedetta Canfora, Rey Audie Escosio, Otilia Boldea, Ana Nunes, Ganna Rozhnova

    Understanding the role of elementary and secondary schools in SARS-CoV-2 transmission is essential for designing effective public health policies. We developed a stochastic age-stratified transmission model, fitted to Dutch epidemiological and contact survey data, to evaluate how alternative closure and reopening strategies for elementary and secondary schools might have shaped the course of the pandemic in the Netherlands from early 2020 through late 2021. Using counterfactual simulations combined with time-varying elasticity analysis, we found that school closures had the potential to mitigate transmission and reduce burden on healthcare. However, their effectiveness was highly context-dependent, influenced by school type, contact patterns, policy timing, and population immunity. Our findings indicate that the contributions of different age groups to transmission shifted over time. Adults were the primary drivers early in the pandemic, followed by adolescents in late 2020 and early 2021, and children in late 2021. These dynamics were reflected in the changing relative impact of different educational levels on overall transmission and hospital burden. Secondary schools had a greater effect on national hospitalizations in 2020, while elementary schools became more important in 2021, partly due to lower prior infection-induced immunity and negligible vaccination coverage among children compared to adolescents. Together, these results highlight the importance of accounting for age-specific transmission dynamics and changing epidemiological landscape when evaluating school closure and reopening strategies. Our study supports a more targeted and adaptive approach to school policies in future pandemics.

  • Open Access

    Testing for multiple change-points in macroeconometrics: an empirical guide and recent developments

    by Otilia Boldea, Alastair R. Hall

    We review recent developments in detecting and estimating multiple change-points in time series models with exogenous and endogenous regressors, panel data models, and factor models. This review differs from others in multiple ways: (1) it focuses on inference about the change-points in slope parameters, rather than in the mean of the dependent variable - the latter being common in the statistical literature; (2) it focuses on detecting - via sequential testing and other methods - multiple change-points, and only discusses one change-point when methods for multiple change-points are not available; (3) it is meant as a practitioner's guide for empirical macroeconomists first, and as a result, it focuses only on the methods derived under the most general assumptions relevant to macroeconomic applications.

  • Open Access

    When do Random Forests work?

    by C. Revelas, O. Boldea, B. J. M. Werker

    We study the effectiveness of randomizing split-directions in random forests. Prior literature has shown that, on the one hand, randomization can reduce variance through decorrelation, and, on the other hand, randomization regularizes and works in low signal-to-noise ratio (SNR) environments. First, we bring together and revisit decorrelation and regularization by presenting a systematic analysis of out-of-sample mean-squared error (MSE) for different SNR scenarios based on commonly-used data-generating processes. We find that variance reduction tends to increase with the SNR and forests outperform bagging when the SNR is low because, in low SNR cases, variance dominates bias for both methods. Second, we show that the effectiveness of randomization is a question that goes beyond the SNR. We present a simulation study with fixed and moderate SNR, in which we examine the effectiveness of randomization for other data characteristics. In particular, we find that (i) randomization can increase bias in the presence of fat tails in the distribution of covariates; (ii) in the presence of irrelevant covariates randomization is ineffective because bias dominates variance; and (iii) when covariates are mutually correlated randomization tends to be effective because variance dominates bias. Beyond randomization, we find that, for both bagging and random forests, bias can be significantly reduced in the presence of correlated covariates. This last finding goes beyond the prevailing view that averaging mostly works by variance reduction. Given that in practice covariates are often correlated, our findings on correlated covariates could open the way for a better understanding of why random forests work well in many applications.


  • Open Access

    Retrospective evaluation of school-related measures on pre-vaccination transmission dynamics of SARS-CoV-2

    by Benedetta Canfora, Rey Audie Escosio, Christiaan H. van Dorp, Otilia Boldea, Marc Bonten, Ana Nunes, Ganna Rozhnova

    Schools are important settings for respiratory virus transmission and a major focus of pandemic control measures in the absence of effective pharmaceutical interventions. During the COVID-19 pandemic, decisions regarding school closures and openings were highly debated due to uncertainties about their population-level impact. Using an age-stratified transmission model formally fitted to hospitalization and seroprevalence data in a Bayesian framework, we retrospectively assessed the effect of school closures and openings on SARS-CoV-2 transmission in Portugal before mass vaccination. We compared the observed epidemic trajectory with two counterfactual scenarios: one in which schools remained open during the first lockdown of spring 2020 and another in which they closed for in-person education following the summer holidays in autumn 2020. We found that keeping schools open during the first lockdown could have led to a worse situation, with hospitalizations rising far beyond observed levels in spring 2020. However, closing schools after the summer holidays alone would not have been sufficient to prevent the autumn wave of hospitalizations in 2020, though it could have mitigated its impact. These findings underscore that the epidemiological impact of school closures is highly context-dependent, influenced by mitigation measures within schools and the intensity of transmission outside them. While closures can support control efforts, especially early into a pandemic, their benefits must be weighed against substantial societal costs. Our results suggest that integrated strategies combining school-based and broader community interventions are essential to effectively reduce transmission during future pandemics.


  • Open Access

    When does Subagging Work?

    by Christos Revelas, Otilia Boldea, Bas J. M. Werker

    We study the effectiveness of subagging, or subsample aggregating, on regression trees, a popular non-parametric method in machine learning. First, we give sufficient conditions for pointwise consistency of trees. We formalize that (i) the bias depends on the diameter of cells, hence trees with few splits tend to be biased, and (ii) the variance depends on the number of observations in cells, hence trees with many splits tend to have large variance. While these statements for bias and variance are known to hold globally in the covariate space, we show that, under some constraints, they are also true locally. Second, we compare the performance of subagging to that of trees across different numbers of splits. We find that (1) for any given number of splits, subagging improves upon a single tree, and (2) this improvement is larger for many splits than it is for few splits. However, (3) a single tree grown at optimal size can outperform subagging if the size of its individual trees is not optimally chosen. This last result goes against common practice of growing large randomized trees to eliminate bias and then averaging to reduce variance.

  • Open Access

    Age-specific transmission dynamics of SARS-CoV-2 during the first 2 years of the pandemic

    by Otilia Boldea, Amir Alipoor, Sen Pei, Jeffrey Shaman, Ganna Rozhnova

    During its first 2 years, the SARS-CoV-2 pandemic manifested as multiple waves shaped by complex interactions between variants of concern, non-pharmaceutical interventions, and the immunological landscape of the population. Understanding how the age-specific epidemiology of SARS-CoV-2 has evolved throughout the pandemic is crucial for informing policy decisions. In this article, we aimed to develop an inference-based modeling approach to reconstruct the burden of true infections and hospital admissions in children, adolescents, and adults over the seven waves of four variants (wild-type, Alpha, Delta, and Omicron BA.1) during the first 2 years of the pandemic, using the Netherlands as the motivating example. We find that reported cases are a considerable underestimate and a generally poor predictor of true infection burden, especially because case reporting differs by age. The contribution of children and adolescents to total infection and hospitalization burden increased with successive variants and was largest during the Omicron BA.1 period. However, the ratio of hospitalizations to infections decreased with each subsequent variant in all age categories. Before the Delta period, almost all infections were primary infections occurring in naive individuals. During the Delta and Omicron BA.1 periods, primary infections were common in children but relatively rare in adults who experienced either reinfections or breakthrough infections. Our approach can be used to understand age-specific epidemiology through successive waves in other countries where random community surveys uncovering true SARS-CoV-2 dynamics are absent but basic surveillance and statistics data are available.


  • Open Access

    Age-specific transmission dynamics of SARS-CoV-2 during the first two years of the pandemic

    by Otilia Boldea, Amir Alipoor, Sen Pei, Jeffrey Shaman, Ganna Rozhnova

    During its first two years, the SARS-CoV-2 pandemic manifested as multiple waves shaped by complex interactions between variants of concern, non-pharmaceutical interventions, and the immunological landscape of the population. Understanding how the age-specific epidemiology of SARS-CoV-2 has evolved throughout the pandemic is crucial for informing policy decisions. We developed an inference-based modelling approach to reconstruct the burden of true infections and hospital admissions in children, adolescents and adults over the seven waves of four variants (wild-type, Alpha, Delta, Omicron BA.1) during the first two years of the pandemic, using the Netherlands as the motivating example. We find that reported cases are a considerable underestimate and a generally poor predictor of true infection burden, especially because case reporting differs by age. The contribution of children and adolescents to total infection and hospitalization burden increased with successive variants and was largest during the Omicron BA.1 period. Before the Delta period, almost all infections were primary infections occurring in naive individuals. During the Delta and Omicron BA.1 periods, primary infections were common in children but relatively rare in adults who experienced either re-infections or breakthrough infections. Our approach can be used to understand age-specific epidemiology through successive waves in other countries where random community surveys uncovering true SARS-CoV-2 dynamics are absent but basic surveillance and statistics data are available.