Scientific Programme

Day Theme: Survey Design in Modern Official Statistics

Adaptation of survey designs to new data sources and changing data environments.

Jeļena Voronova (Central Statistical Bureau of Latvia, representative of Latvia in BNU Network)

Jānis Valeinis (Laboratory of Statistical Research and Data Analysis, University of Latvia)

Raimonds Lapiņš (Central Statistical Bureau of Latvia)

Session 1

Keynote lectures

Chair: Tomasz Żądło

Frauke Kreuter (Ludwig-Maximilians-University of Munich) 

Improving Official Statistics with AI, and AI with Official Statistics

Danutė Krapavickaitė (Vilnius Gediminas Technical University)

Historical Aspects of the Establishment of the BNU Network

Andris Fisenko (Latvijas Banka)

Memories of the BNU Network: A Historical Photo Gallery

Session 2

Contributed presentations with discussions

Chair: Vilma Nekrašaitė-Liegė

Uldis Ainārs (Central Statistical Bureau of Latvia) 

Where Survey Design Now Begins and Ends and Is It Still a Survey at All? 

Kristīne Gruzinska (Central Statistical Bureau of Latvia)  

Use of Data from Administrative Data Sources of Legal Entities to Reduce Respondent Burden in the Household Budget Survey 

Discussant: Ieva Burakauskaitė Ramašauskienė

Agris Raipalis (University of Latvia) 

Methodological Framework for Expert Selection in Multi-Sectoral Cross-Border Research: Evidence from the Baltic States 

Day Theme: Machine Learning in Data Preparation and Analysis

Use of machine learning methods to improve data cleaning, integration, and statistical analysis.

Session 3

Keynote lectures

Chair: Jānis Valeinis

Marco Puts (Statistics Netherlands)

Total Machine Learning Error Model

Anastasija Tetereva (Erasmus University Rotterdam)

Tree-Based Methods for Survey Data and Beyond: Modeling Structured Heterogeneity with Interpretable Machine Learning

Session 4

Contributed presentations with discussions

Chair: Maria Valaste

Oļegs Krasnopjorovs (Latvijas Banka, University of Latvia)  

Quantifying Quality of Life in the European Cities Using Eurobarometer Survey Data 

Valerie Schilting (University of Helsinki)  

Machine Learning for Missing Data: An Overview of the MICE Algorithm 

Oļesja Nikoluškina (Central Statistical Bureau of Latvia)  

Imputation Methods for Construction Statistics 

Anastasija Meļņičuka (Central Statistical Bureau of Latvia)  

Latvian Household Budget Survey: Post-Fieldwork Statistical Processing 

Marta Tropa (University of Latvia)  

Multimodal Machine Learning: Combining Multiple Data Modalities 

Session 5

Invited lectures

Chair: Jānis Valeinis

Olena Mulyk (Igor Sikorsky Kyiv Polytechnic Institute)  

Monitoring PTSD Factors in Women of Reproductive Age Using Supervised and Unsupervised Machine Learning Methods (online) 

Rihards Roberts Miķelsons (University of Latvia)  

Development of a Machine Learning Model for Identifying Deficiencies in Family Nucleus Labels 

Tetiana Makedon (Taras Shevchenko National University of Kyiv) 

Developing a Composite Wartime Urban Comfort Index for Ukraine Through the Fusion of Web-scraped Economic Data and Open-source Air Raid Alert Statistics (online) 

Rafael Zimmerle (University of Perugia) 

Stability in Model-Assisted Estimation 

Group photo; guided tour of the House of Science

Session 6

Contributed presentations with discussions

Chair: Leonora Pahirko

Vilma Nekrašaitė-Liegė (Vilnius Gediminas technical University)  

Teaching Calibration Estimation Within a Modern Survey Sampling Curriculum  

Akvilė Vitkauskaitė (Vilnius University, State Data Agency (Statistics Lithuania))

Estimation of Population Parameters with Partial Social Media Coverage  

Biruta Sloka (University of Latvia)  

Use of Official Statistics Survey’s Micro Data in Studies and Research  

Hanna Livinska (Taras Shevchenko National University of Kyiv)  

Next-Generation Survey Analytics in Social Data Science Education 

Day Theme: Probability and Non-Probability Sampling

Methods for integrating different data sources to produce reliable and efficient statistical estimates from both probability and non-probability data.

Session 7

Invited lectures

Chair: Biruta Sloka

Tomasz Żądło (University of Economics in Katowice) 

Fundamentals and Recent Developments in Small Area Estimation 

Session 8 

Contributed presentations with discussions 

Chair: Biruta Sloka 

Liana Radeckaja (Vilnius Gediminas Technical University) 

Development of Digital Learning Tasks for Teaching Small Area Estimation in Sampling Methods Courses

Anželika Ņesterova (Faculty of Science and Technology, University of Latvia) 

Small Area Estimation for Household Budget Survey: Producing Reliable Estimates for Municipal Domains 

Session 9

Invited lecture

Chair: Jeļena Voronova

Volodymir Sarioglo (Institute for Demography and Social Studies) 

Population Sample Surveys in Ukraine During Wartime: Challenges and Lessons

Session 10 

Contributed presentations with discussions 

Chair: Jeļena Voronova

Ieva Strēle (Riga Stradins University)  

Agreement Between Healthcare Service Data and the Diabetes Register in Latvia

Heilika Nigulas (University of Tartu)  

Selection Bias and Methods for its Correction: The Case of the Estonian Biobank

Marwan Babiker (University of Latvia)

Determinants of Online Survey Response Among Primary Healthcare Workers: a Full Workforce Evaluation of Design and Procedural Factors (online)

Meeting with the University of Latvia Leadership

Round table discussions

Day Theme: Small Area Estimation and Nonparametric Approaches

Development of detailed statistics using small-area models and classical, robust, and nonparametric statistical methods.

Session 11

Invited lectures

Chair: Māra Delesa-Vēliņa

Danutė Krapavickaitė (Vilnius Gediminas Technical University) 

Integration of Non-Probability and Probability Samples 

Marcin Szymkowiak (Poznan University of Economics and Business) 

A Joint Calibration Approach for Totals and Quantiles for Probability and Non-Probability Samples 

Session 12

Contributed presentations with discussions 

Chair: Ieva Strēle 

Tetiana Ianevych (Taras Shevchenko National University of Kyiv)  

Integrating Information from Non-Probability Samples: Case Study

Nataliia Korzhunova (Mykhailo Ptoukha Institute for Demography and Life Quality Research of the National Academy of Sciences of Ukraine)  

Beyond Traditional Sample Surveys: Multi-Source Data Integration for Labour Market Research in Ukraine

Liliāna Roze (Central Statistical Bureau of Latvia)  

Income Imputation Challenges in Labour Force Survey: Methods and Practical Issue

Ieva Burakauskaitė-Ramašauskienė (Vilnius University, State Data Agency (Statistics Lithuania))

Making Valid Population Inferences from Panel-based Survey Data

Discussant: Joonas Sova

Session 13

Invited lecture

Chair: Danutė Krapavickaitė 

Kaja Sõstra (Statistics Estonia, EUROSTAT)  

Developing Transparent and Harmonised Publication Thresholds for EU-LFS Data Dissemination

Session 14 

Contributed presentations with discussions 

Chair: Danutė Krapavickaitė 

Kristi Lehto (Statistics Estonia)  

Revision of the calculation of the Publication Thresholds for Estonian Labour Force Survey

Thomas Laitila (Örebro University)

On Non-Response Follow Up Studies

Session 15

Contributed presentations with discussions 

Chair: Henri Luomaranta-Helmivuo

Ēriks Vilunas (University of Latvia) 

Potential of robust and nonparametric methods in survey sampling (TBC)

Olga Vasylyk, Oksana Lahoda (Igor Sikorsky Kyiv Polytechnic Institute) 

Application of Bootstrap Method to Insurance Data (online) 

Emīls Siliņš (University of Latvia)  

Non-parametric Covariate Adjustment Methods for Statistical Inference (online) 

Sofiia Lukashevych (University of Latvia) 

Empirical Likelihood Residual Bootstrap Method

Discussant: Tetiana Ianevych

Day Theme: AI and Automation in Statistical Production

Application of AI and automation to improve the efficiency and quality of statistical production processes.

Session 16

Invited lectures  

Chair: Thomas Laitila 

Henri Luomaranta-Helmivuo (Statistics Finland)  

Adaptive Survey Improvements - AI Meets Survey Science 

Marco Puts (Statistics Netherlands) 

Adopting AI in Official Statistics: The Challenges Ahead 

Session 17 

Contributed presentations with discussions 

Chair: Kaja Sõstra

Maria Valaste (University of Helsinki, Centre for Social Data Science) 

From Burden to Backbone: Large Language Models for Survey Data Documentation 

Oļesja Larionova (Central Statistical Bureau of Latvia) 

A Practical Approach to COICOP Classification of Scanner Data Using LLMs and Embeddings 

Dariia Drozd (University of Latvia)  

Sampling and Nonresponse Challenges in Measuring AI-enabled Financial Management 

Maksim Čižov (State Data Agency, Vilnius University) 

Application of Data Synthesis Methods for Opening Business Microdat