1 item has been added to your cart.
Join us in Cluj-Napoca, Romania for the first Carpathian Stata Conference at Universitatea Babeș-Bolyai!
Meet researchers from different disciplines, discover new applications highlighting Stata’s potential capabilities for applied research, exchange new community-contributed commands developed for Stata, and interact directly with statisticians from StataCorp. There will be an optional workshop on 28 July.
Considered the unofficial capital of Transylvania, Cluj-Napoca is home to a diverse cultural scene, vibrant nightlife, and landmarks such as Saint Michael's church, the Palace of Justice, and a collection of theaters and museums.
All times are EEST (UTC +3)
| 8:35–8:45 | Opening Remarks and the history of Wald |
Abraham Wald Keynote Lecture |
|
| 8:45–9:45 | Generalized method of moments estimation of linear dynamic panel-data models
Abstract:
In dynamic models with unobserved group-specific effects, the lagged dependent variable is an endogenous regressor by construction. The conventional fixed-effects estimator is biased and inconsistent under fixed-T asymptotics. To deal with this problem, “difference GMM” and “system GMM” estimators are predominantly applied in practice. I discuss recent developments in this area and present the Stata package xtdpdgmm, which provides a lot of flexibility in specifying the estimator—including nonlinear moment conditions, forward-orthogonal deviations, iterated or continuously updated GMM, and doubly robust standard errors. Useful postestimation features include overidentification and underidentification tests, as well as newly proposed serial-correlation tests.
Sebastian Kripfganz
University of Exeter
About Abraham Wald:
Abraham Wald was a preeminent researcher born in Cluj in 1902, whose work in the fields of geometry, probability, and mathematical economics informed and shaped the work of researchers in Cluj and across the world. Learn more about Abraham Wald.
|
| 9:45–10:00 | Regression with observational multilayered network data
Abstract:
A novel method to estimate social effect coefficients in the popular so-called linear-in-means regression model in the social sciences is presented here that utilizes nonexperimental multidimensional network data. The procedure can accommodate social interactions that correlate with the error in the model by making use of a different set of network links among the same observations that are exogenous in the traditional sense. In particular, the full observability of a two-layered multiplex network data structure is assumed here to propose a new generalized three-stage least-squares (G3SLS) estimator that is consistent, asymptotically normally distributed, and also easy to implement using widely-used existing statistical software because of its closed-form definition. The underlying assumptions are general enough to accommodate common problems with observational data such as measurement error, simultaneity, and unobserved heterogeneity. Monte Carlo exercises confirm the good small-sample performance of the proposed G3SLS estimator in these scenarios. An empirical application finds positive and significant peer effects in citations among research articles published in top general-interest journals in economics.
Kim Huynh
Indiana University and Laboratoire d’Économie d’Orléans
|
| 10:00–10:10 | Break |
| 10:10–10:35 | Heterogeneous effects of sanctions on institutional quality in target states
Abstract:
The effectiveness of sanctions has been the subject of intense debate over the last few years, but their impact on institutional quality in target states remains largely unexplored. This study uses quantile regression to assess the impact of international sanctions imposed by the US, the EU, and the UN on institutional quality. I use data for 194 countries from 1990 to 2023 and distinguish between economic and noneconomic sanctions. My results show that sanctions have a detrimental impact on institutional quality in the targeted countries. The impact is quite heterogeneous and depends on both the type of sanction imposed and the sender. In the vast majority of cases, better institutional quality in the target state can partially alleviate the negative impact of sanctions, the intensity decreasing with institutional quality. From a policy perspective, my study highlights that sanctions impose unintended costs on target countries by weakening their capacity to strengthen their institutional systems.
Irina-Marilena Ban
Universitatea Babeș-Bolyai
|
| 10:35–11:00 | Modeling the adoption and fiscal effects of Romania’s 2024 tax amnesty
Abstract:
Romania's 2024 fiscal amnesty offers a timely case for examining how debt-relief policies are implemented across local public authorities and whether they improve short-run fiscal collection. The national measure allowed eligible individual debtors to obtain partial cancellation of outstanding budget obligations conditional on repayment. For obligations owed to local budgets, however, application depended on decisions adopted by local councils. This creates variation in implementation across administrative-territorial units. This presentation uses Stata to construct a municipality-period panel, code local council adoption dates, and estimate the determinants of implementation using discrete-time event-history models. Explanatory variables include local fiscal pressure, own-source revenue, arrears, population size, socioeconomic indicators, and political alignment. The second part of the presentation compares local fiscal outcomes before and after implementation, focusing on changes in tax collection and arrears around the adoption date. The presentation contributes early evidence on how Romania's 2024 fiscal amnesty was implemented locally and whether it was associated with improved fiscal collection.
Florina Burdet
Universitatea Babeș-Bolyai
|
| 11:00–11:10 | Break |
| 11:10–11:35 | Buy, buy, bye! The role of learning and beliefs in exiting from the Bitcoin market
Abstract:
I build and test a theoretical model of Bitcoin adoption and unadoption in which beliefs about Bitcoin’s long-run survival and heterogeneous learning rates drive entry, exit, and reentry decisions. The model implies ranked belief distributions across never owners, owners, and past owners and describes how these distributions and the associated ownership shares evolve over time. Using 2017–2019 data from the Bank of Canada’s Bitcoin Omnibus Survey and a sequential logit model with control functions, the empirical belief distributions, which are a key determinant of both entry and exit, align with the model: owners are the most optimistic, never owners the most pessimistic, and past owners lie in between, with past owners becoming more optimistic over time. A composite cryptofinancial literacy (CFL) index plays a context-dependent role, because high-CFL individuals are less likely to exit during downturns but more likely to exit when market conditions improve, while medium- and low-CFL individuals are less likely to enter yet more inclined to exit. Counterfactual simulations that shift the population distribution of CFL upward show that higher CFL increases both current and past ownership, especially among respondents with strong survival beliefs, highlighting a complementarity between literacy and optimism in shaping the Bitcoin adoption lifecycle.
Daniela Balutel
National Bank of Moldova and Alexandru Ioan Cuza University
|
| 11:35–12:00 | Household adjustment to economic shocks: Saving, consumption smoothing, and life satisfaction
Abstract:
This presentation examines how households adjust to large economic shocks and how these responses shape life satisfaction. Using panel data from the Slovak household finance and consumption survey combined with district-level variation in COVID-19 severity, the analysis distinguishes between exposure to the shock, household adjustment, and the transmission of these adjustments into life satisfaction. Empirically, the presentation combines a difference-in-differences design with a mediation framework to examine how shock severity affected income, consumption, wealth, and saving behavior and how these changes were reflected in life satisfaction. The results reveal substantial and uneven economic disruption beneath stable average life satisfaction. Income losses and consumption adjustments are concentrated toward the upper end of the preshock income and consumption distributions, indicating marked heterogeneity in exposure. Households most affected by income losses adjust through saving decumulation and reductions in discretionary consumption, while basic consumption remains comparatively stable. These patterns are consistent with intertemporal consumption smoothing. Despite substantial economic adjustment, the transmission of these changes into life satisfaction is limited. The resulting indirect effects operating through income, consumption, and wealth are small and statistically insignificant. The findings show that the absence of an average effect reflects attenuated transmission rather than limited economic disruption. More broadly, the results underscore that the welfare consequences of large shocks depend not only on exposure but also on the mechanisms through which households adjust and absorb disruption.
Peter Tóth
Národná banka Slovenska and Ekonomická univerzita v Bratislave
|
| 12:00–1:00 | Lunch |
| 1:00–2:00 | Heterogeneous DID when units switch in and out of treatment
Abstract:
In this talk, I will introduce the new xtswitchdid command, which provides event-study treatment effects for panel data when subjects are allowed to switch in and out of treatment. This is an implementation of the estimator proposed in de Chaisemartin and Xavier 'D'Haultfœuille (2024). I will also discuss how xtswitchdid fits into the DID estimators that have been implemented in the past couple of Stata releases and how it fits the evolution of our understanding of DID.
Reference: Chaisemartin, Clément de, Xavier D’Haultfœuille, and Gonzalo Vazquez-Bare. 2024. Difference-in-difference estimators with continuous rreatments and no stayers. AEA Papers and Proceedings 114 (May 2024): 610–13. https://doi.org/10.1257/pandp.20241049.
Enrique Pinzón
StataCorp
|
| 2:00–2:15 | Break |
| 2:15–2:40 | Double/debiased machine learning for conditional average treatment-effect estimation
Abstract:
I compare double/debiased machine learning (DDML) for estimating conditional average treatment effects (CATEs) in Stata and Python. The focus is practical: what Stata users can do directly with the community-contributed ddml package, official Stata CATE tools, and related commands; where Python or R can add value; and how these environments can be combined in applied work. I illustrate how partially linear and interactive DDML models can be used to estimate heterogeneous treatment effects, reducing regularization and overfitting bias through orthogonalization and cross-fitting. I then compare alternative approaches to estimating the treatment-effect function itself, including structured heterogeneous PLM specifications, R-learners, DR-learners, and causal forests. Empirical illustrations use random-assignment data from the Job Training Partnership Act (JTPA) and the National Supported Work (NSW) demonstration, as well as census-based wage applications. I emphasize that machine learning can improve the estimation of control functions and treatment effects, but this does not replace credible identification or careful choices about learners, tuning parameters, and interpretability. I argue that Stata is currently especially useful for transparent DDML estimation and reporting of heterogeneous treatment effects, while Python and R can be useful complements for implementing a broader set of more flexible DDML-CATE meta-learners.
Miana Plesca
University of Guelph
|
| 2:40–3:05 | Do environmental and energy expenditures pay off? Evidence on firm-level productivity in Romania
Abstract:
This study explores the effect of environmental protection effort, both the adoption and the size of environmental protection expenditure, as well as the energy dependence on total factor productivity (TFP) at the firm level. To this end, I focus on an extended sample of Romanian firms observed during the period 2020–2023. The results, consistent across different specifications, indicate that firms engaged in environmental protection activities have, on average, higher TFP than similar firms but that environmental protection expenditures induce additional costs, negatively influencing TFP. Moreover, energy intensity has a positive and significant effect on TFP. Overall, these findings provide important insights into the relationship between these indicators at the firm level and may lead to the development and improvement of more tailored associated policies.
Alexandra-Anca Purcel
Universitatea Babeș-Bolyai
|
| 3:05–3:30 | Climate-related communication of central banks and bank risk
Abstract:
This study examines the influence of green rhetoric from central banks on banking institutions. Using a sample of 437 publicly listed banks from 43 countries between 2000 and 2019, I find a positive correlation between an increased proportion of climate-related discourse in central bank speeches and a reduction in both systematic and systemic risk for banks. This may be attributed to improved transparent communication by central banks, which reduces individual and systemic risk for banks. This, in turn, supports the accountability and independence of central banks, which are negatively correlated with bank risk-taking behavior. In a similar vein, central banks that are most vocal about climate issues are also leading the way in adopting climate-related policies that facilitate the transition to net zero, which are positively associated with financial stability. The findings of this study have significant policy implications in the context of central bank's growing involvement in climate-related issues and their consequent shaping of market participants' perceptions.
Nicu Sprincean
Universitatea „Alexandru Ioan Cuza“
|
| 3:30–3:45 | Break |
| 3:45–4:10 | Signals from the Web: What Google trends reveals in payment digitization
Abstract:
Using Google Search Volumes (GSV), I develop the GSV-based Digitization of Payments Index (GSVDPI), a multidimensional metric built via machine and deep learning models, to track the evolution of digital payments. This index allows for quantifying the changes in the use of digital payment instruments (card, contactless, mobile, and others) in France and comparison with other European countries like Germany, Italy, and Spain. Compared with standard payment digitization statistics published by institutions like the European Central Bank (ECB), the GSVDPI offers distinctly valuable and real-time insights. Consequently, the GSVDPI accounts for the complex dynamics and policy expectations of the digitization process, which are typically overlooked by univariate metrics. I contribute to the literature by identifying the economic mechanisms driving digital payment adoption and examining its macroeconomic effects through a cross-country analysis. I show that incorporating the GSVDPI into the explanatory variables of a multivariate cash demand framework significantly enhances model performance, thereby offering deeper insights into future payment trends.
Louis-Alexandre Bayol
Banque de France
|
| 4:10–4:35 | Uneven passthrough of VAT cuts to food prices: Evidence from scanner data
Abstract:
This presentation studies how consumer prices responded to two permanent VAT reductions in Slovakia, both lowering the rate from 20% to 10%, first for essential staples in 2016 and later for a broader set of goods in 2020. Leveraging detailed scanner data and a synthetic difference-in-differences framework, I find that VAT passthrough is highly heterogeneous depending on product attributes, demand elasticity, and policy design. The 2016 reform, which targeted clearly defined essential goods and was announced well in advance, led to full and persistent price reductions. The 2020 reform, applied to a loosely defined category of healthy goods with limited legislative lead time, produced only partial and short-lived price effects, with passthrough largely reversing within six weeks. These contrasting outcomes identify policy design as a key determinant of passthrough. Well-targeted VAT cuts on essential goods with inelastic demand can deliver meaningful and durable consumer price relief, while broader or less transparent interventions risk increasing retailer margins without lasting benefits for consumers.
Contributors:
Brian Fabo
Národná banka Slovenska and Slovenská akadémia vied
Peter Tóth
Národná banka Slovenska and Ekonomická univerzita v Bratislave
Pavel Gertler
Národná banka Slovenska
|
| 4:35–4:50 | Break |
| 4:50–5:15 | Teaching with self-made help files
Abstract:
Teaching statistics using Stata in a scientific way should involve the usage of Stata programming language instead of only clicking single commands from the pull-down menus. Like in any other language, students very often make small spelling mistakes that cause error messages, which are frustrating at the beginning. The concentration on these formal issues often enough keeps them from intuitively understanding commands' constructive logic. In the presentation, I would like to show how teachers can make use of Stata's markup language (SMCL) in order to construct help files, where you can add executable Stata commands. Students are able to follow the commands' logic by simply clicking on them and get a better feeling of the command structure before they have to care about spelling issues. I will demonstrate the construction and the usage of such help files by showing examples of graph commands, which usually have a lot of options, and the relatively new etable command, which is made for table export.
Christian Brzinsky-Fay
DPC Software
|
| 5:15–5:30 | Break |
| 5:30–6:00 | Open panel discussion with Stata developers
Contribute to the Stata community by sharing your feedback with StataCorp's developers. From feature improvements to bug fixes and new ways to analyze data, we want to hear how Stata can be made better for our users.
|
Sebastian Kripfganz
University of Exeter
In dynamic models with unobserved group-specific effects, the lagged dependent variable is an endogenous regressor by construction. The conventional fixed-effects estimator is biased and inconsistent under fixed-T asymptotics. To deal with this problem, “difference GMM” and “system GMM” estimators are predominantly applied in practice. I discuss recent developments in this area and present the Stata package xtdpdgmm, which provides a lot of flexibility in specifying the estimator—including nonlinear moment conditions, forward-orthogonal deviations, iterated or continuously—updated GMM, and doubly-robust standard errors. Useful postestimation features include overidentification and underidentification tests, as well as newly proposed serial-correlation tests.
Enrique Pinzón
StataCorp
In this talk, I will introduce the new xtswitchdid command, which provides event-study treatment effects for panel data when subjects are allowed to switch in and out of treatment. This is an implementation of the estimator proposed in de Chaisemartin and Xavier 'D'Haultfoeuille (2024).
I will also discuss how xtswitchdid fits into the DID estimators that have been implemented in the past couple of Stata releases and how it fits the evolution of our understanding of DID.
Christian Brzinsky-Fay
DPC Software GMBH
Teaching statistics using Stata in a scientific way should involve the usage of Stata programming language instead of only clicking single commands from the pull-down menus. Like in any other language, students very often make small spelling mistakes that cause error messages, which are frustrating at the beginning. The concentration on these formal issues often enough keeps them from intuitively understand the commands’ constructive logic.
In the presentation, I would like to show how teachers can make use of Stata’s markup language (SMCL) in order to construct help files, in which you are able to construct help files, where you can add executable Stata commands. Students are able to follow the commands logic by simply clicking on them and get a better feeling of the command structure before they have to care about spelling issues.
I will demonstrate the construction and the usage of such help files by showing examples of graph commands, which usually have a lot of options, and the relatively new etable command, which is made for table export.
Enrique Pinzón, Director of Econometrics, StataCorp
28 July 2026
This course covers most of the estimators for causal inference available in Stata. We will discuss the following:
We will also present some theoretical results along with worked Stata examples.
The conference will be hosted at Universitatea Babeș-Bolyai, Faculty of Economics and Business Administration. Participants are asked to travel at their own expense. The conference fees include admission to all scientific sessions, materials, breaks, lunches, and dinners.
| Conference fees VAT not incl. |
Student | Other |
|---|---|---|
| Conference | € 25 | € 35 |
| Conference + workshop | € 50 | € 75 |
To ensure affordability and ease of purchase, our official distributor, TStat S.r.l, will accept and process registrations for the conference.
The registration deadline is 22 July 2026.
StataCorp is the logistics organizer for the 2026 Carpathian Stata Conference. The co-organizers are TStat S.r.l., DPC Software GmbH, and Token Communication SRL—the official distributors of Stata in this region and the surrounding regions—and Universitatea Babeș-Bolyai, our host for the conference.
View the proceedings of previous Stata Conferences and international meetings.
