Seminars

Spatial Warsaw Seminar

Aleksandar Petreski, Ph.D. | Housing market segmentation and persistence in idiosyncratic risk: spatial-temporal GARCH at transactions level
Aleksandar Petreski, Ph.D., Jönköping International Business School, Jönköping University, Sweden | February 21st, 2024

This research, using spatial-temporal framework, analyzes if segmentation of housing market by object type (houses vs apartments), affects the features of idiosyncratic volatility pattern: volatility clustering, volatility mean reversion and asymmetry. Volatility features are tested at transaction level by novel spatial – temporal ARCH and GARCH model, applied to unbalanced non-repetitive sales data from Jönköping, Sweden, from 2012 through 2019. Results show significant difference in the price volatility patterns between the market segment of apartments and houses. Houses have higher constant price variance. With wider geographical distances between observations, constant price variance of houses increases, as more heterogeneous areas are included. Estimated parameters for the segment of apartments show positive shock response and positive memory effect, which confirm the assumption of spatial-temporal persistence in the variance. Apartments show significantly higher response to negative price shocks than to the positive price shocks. This asymmetry is clearly observed in the spatial context compared to the spatial-temporal one. Spatial-temporal models have better goodness-of-fit and better statistical adequacy compared with the benchmark spatial ARCH model.

Alberto Hidalgo, MA | Your Room is Ready: Tourism and Urban Revival

Alberto Hidalgo, MA, IMT School for Advanced Studies Lucca | March 20th, 2024

Tourism is an essential sector of the global economy, contributing significantly to GDP and employment. Despite its importance, our understanding of its impact on urban economic activity remains limited. This paper aims to fill this gap by examining the impact of tourism on urban transformation using a dataset of hotel openings in Madrid from 2001-2010. I show that hotel openings positively impact the number of establishments and employment by using the number of protected buildings as an instrumental variable to account for the non-random distribution of hotel openings. Interestingly, hotel openings contribute to changes in the composition of the economic activities and the business structures, enhancing tourist-oriented corporate-owned businesses over other individual-owned companies. Finally, economic effects extend to the real estate market, increasing rental prices and residential investment.

Maria Kubara, MA, | Unravelling causal CBD impact on technological startup growth

Maria Kubara, MA, Faculty of Economic Sciences, University of Warsaw | April 17th, 2024

This study investigated the impact of the Central Business District (CBD) location on the growth trajectories of technological startups, utilizing the causal forest methodology to analyse the heterogeneous treatment effects at the micro-geographical level. This analysis reveals that CBD’s influence on startups largely depends on their initial capital endowments, with well-resourced firms experiencing significant benefits. However, CBD’s location becomes less beneficial as startups expand their employment size. This research provides insights into the effects of intra-urban location decisions and methodological advancements in regional science, demonstrating the usability and efficiency of causal forest for estimating Heterogenous Treatment Effects. Causal forest allows the entanglement of complex, non-linear relations between the company’s initial conditions and the urban ecosystem, providing a reliable and robust framework for future studies on causal effects on the micro-geographical scale. This dual contribution enriches the literature on the impact of business location on individual firms and provides a framework for future studies focused on causal relations in urban studies.

Zehra Usta, MA | Spatio-Temporal Evolution of Patent Clusters: A Comprehensive Analysis from 1980 to 2010 in the UK

Zehra Usta, MA, Faculty of Economic Sciences, University of Warsaw | April 17th, 2024

This research investigates the spatio-temporal changes of patent data clustering, examining shifts in their spatial distributions over time. The study explores the dynamic nature of patent clusters and their changes across different periods. Furthermore, the research explores coexistence among different patent types, unveiling relationships within and between patent clusters. By employing unsupervised learning methods, this study contributes to a comprehensive understanding of the changing structure of patent clusters, offering insights into the changing dynamics and interconnections within patent clusters across different time periods. The study will verify that present-day clusters are displaying greater diversity compared to those in the past. This hypothesis will be verified on the basis of very uncommon data on exact locations of patents which will be a significant improvement compared to the previous studies which are using regional data instead of point data.

Carles Méndez-Ortega | Location Patterns and Drivers of Coworking Spaces in European Regions

Carles Méndez-Ortega | Faculty of Economics and Business, Universitat Oberta de Catalunya | May 15th, 2024

The location patterns of new working spaces and the factors driving their emergence have been widely studied in Europe. However, research has mainly focused on specific countries and very local areas. We widen this perspective to a regional one and conduct a cross-country study of the location patterns and determinants of today’s main new working spaces, specifically coworking spaces that facilitate collaboration and knowledge sharing. Coworking spaces are traditionally associated in the literature with vibrant urban places, but they are also found in non-metropolitan regions. Drawing on a unique data set of coworking space locations in four large European countries (Spain, France, Germany and Poland), we test how demographic, economic, structural, technological and accessibility-related determinants affect the location of these spaces. We then investigate whether the identified determinants vary between metropolitan and non-metropolitan and between countries, controlling for spatial effects.

Balázs Lengyel | Barriers of Urban Mobility
Head | ANET – Agglomeration and Social Networks Research Lab
Senior Research Fellow | Centre for Economic and Regional Studies, Budapest, Hungary
Research Fellow | NeTI Lab, Centre for Advanced Studies, Budapest Corvinus University

Administrative boundaries, natural obstacles, railways or major roads may contribute to the segregation of neighborhoods. However, the empirical knowledge about this issue is limited. Here, we present a methodological framework to assess the importance of barriers to urban mobility along their hierarchy, over time and across neighborhoods. Using GPS mobility data, we construct a network of blocks from the sequence of individual visits in a major European city. A community finding algorithm allows us to evaluate the fit of the neighborhoods detected in this mobility network to characteristic urban barriers by adjusting the resolution parameter of the Louvain method to decrease the dominance of mobility hubs. Using the Symmetric Area Difference index, which quantifies the overlap between the polygons separated by urban barriers and the network communities, we find that the match of detected neighborhoods and urban barriers improves as the resolution parameter increases. However, at high values of the resolution parameter, this fit gets better for lower rank administrative or road barriers than for their higher rank pairs. Although the role of urban barriers in hindering mobility became stronger over the COVID-19 restrictions, their fit with detected neighborhoods deteriorated for most analyzed barriers. Exceptions are primary roads, to which the fit of neighborhoods is stable over periods. By increasing the resolution parameter, we find emergent communities of local mobility flows that cross primary roads and a consequent optimal value of the resolution parameter. The Barrier Crossing Ratio, the fraction of barrier crossings that are network community crossings as well, demonstrates that the barriers’ impact on mobility are significantly different across urban areas. These results contribute to the ongoing discourse on urban segregation, emphasizing the importance of barriers to urban mobility in shaping neighborhood interactions and mixing.

Kateryna Zabarina, Muhammad Usman | Spatial and Temporal Assessment of Conflicts, Violence and Political Demonstrations: Case of Pakistan
Kateryna Zabarina, Muhammad Usman, Faculty of Economic Sciences | June 12th, 2024

Political violence is a complex and multifaceted phenomenon that can have profound consequences for individuals, communities, and entire nations. It is essential for policymakers, scholars, and citizens to comprehend the causes and effects of political violence. While the locations of conflicts and political violence are well known, a more comprehensive and thorough investigation into the spatio- temporal changes of conflicts remains an essential frontier in big data driven conflict studies.
Already known methods of analysis for political violence, conflicts and demonstrations data (both spatial and spatio- temporal) include regression analysis, spatial count modelling and finally, point pattern analysis. The existing literature on conflict studies have summarised the prevailing socioeconomic inequalities and climate change factors as the potential drivers of conflicts.
This study examines the political violence at the regional level in Pakistan from the time period 2010- 2023 using the big data on conflicts provided by Armed Conflict Location & Event Data Project (ACLED). In particular, we use data on four different types of violence, which are battles, riots, explosions and violence against civilians. We combine several approaches, such as point pattern analysis, machine learning and spatial econometrics technics.
Firstly, we discover inseparability of spatial and temporal dimensions. Secondly, our results exhibit strong spatial and temporal effects in four spatial typologies of political violence regions have emerged as the new spots for political violence in recent years. Furthermore, we found the spatial coexistence of riots, battles and remote violence in the northern and western region of Pakistan which are labelled as the hotspots for conflicts. Finally, the 3d plot of spatio-temporal Ripley’s K shows that with increase of both spatial and temporal dimensions, the level of clustering in the analysed point pattern increases.
Obtained results will be useful from several perspectives with wide range of policy implications.

Prof. Carolina Guevara Rosero | Regional disparities in amenities and the life satisfaction of internal migrants
Prof. Carolina Guevara Rosero, Universidad de las Américas (UDLA) | October 14th, 2024

Carolina Guevara obtained her PhD at the Université Jean Monnet de Saint Etienne in France. She is currently the President of the RSAI Ecuadorian Section and associated professor at Universidad de las Américas. She earned the Peter Nijkamp Research Encouragement Award granted by RSAI in 2022. She is positioned as one of the 15 most prominent researchers in Economics, being the first woman in the REPEC ranking. Carolina’s research focusses on productivity, innovation, firm location in Ecuador and more broadly in Latin America, regional and Urban Economics, Geographical Economics, Welfare Economics, Economics of Innovation and Knowledge. Her work is noteworthy for its focus on applied topics with clear policy implications for developing countries in their efforts to achieve higher living standards.

Zofia Bednarowska-Michaiel, PhD | Spatial Inequity in Traveling To Work

Zofia Bednarowska-Michaiel, PhD | Postdoctoral Scholar in the Department of Urban Studies and Planning, University of California, San Diego | Research Labs affiliations: Homelessness Hub and Design Lab | October 28th, 2024

Are London cycle tracks and lanes white lanes? Drawing on regional science, mobilities paradigm, and spatial econometric modeling, I examine cycling inequity using London as a case study.
In my talk, I will present findings from my paper, “Ethnic Inequalities in Cycling to Work in London: Mobilities Injustice and Regional Approach,” published in Regional Studies, Regional Science. I mapped and analyzed ethnic disparities in cycling, showing that living in more ethnically diverse areas reduces the likelihood of commuting by bike. Following this, a research grant from the University of Warwick allowed further exploration of these findings. This project aims to bring forward the quantitative results and identify relevant policies and advocacy strategies to address these inequities.

IDUB Seminar

Prof. Jorge Mateu | A data science journey through the analysis of spatio-temporal point pattern data.

Prof. Jorge Mateu, Department of Mathematics, University Jaume I of Castellon, Spain

Prof. Jorge Mateu conducted a series of lectures which offered a deep dive into spatio-temporal point process statistics. Through lectures and hands-on R sessions, it explored non-parametric intensity estimation, mechanistic vs. empirical models, neural kernel-based processes, and advanced statistical learning, providing comprehensive insights into spatial growth models and data dimensionality reduction.

Prof. Christopher Brunsdon | Spatial Data Handling and Modelling Using R

Prof. Christopher Brunsdon, Director of the National Centre for Geocomputation, Maynooth University, Ireland

Prof. Christopher Brunsdon conducted 15 hours of lectures covering different aspects of spatial data analytics in R, including trip modeling, regression, Markov models, Bayesian methods, and real-world computing challenges, offering insights into spatial algorithms and analytics hurdles.

Nema Dean, Ph.D. | A spatial data science approach to model-based clustering and semi-supervised variable selection

Nema Dean, Ph.D., School of Mathematics & Statistics, University of Glasgow, United Kingdom

Dr Nema Dean conducted a 30 hour course which was focused on the spatial clustering challenges. Her course delved into clustering and spatial modeling with R, covering non-parametric and parametric techniques for data organization. Students have learned about cluster comparison and Bayesian CAR models, gaining skills to analyze spatial data and tackle real-world problems in diverse fields.

Kevin Credit, Ph.D. | Causal Machine Learning for Spatial Data

Kevin Credit, Ph.D., Assistant Professor, National Centre for Geocomputation, Maynooth University

Dr Kevin Credit conducted a 15-hour hands-on course on applying machine learning (ML) to social science topics in spatial context, emphasizing urban studies like transportation and urban development. The course focused on causal ML and spatial data analysis, illustrated through a case study on Chicago’s 606 elevated path and its impact on local development and emissions. Over five sessions, participants engaged in practical exercises using R, building a foundational understanding of ML’s theoretical and methodological aspects in social sciences.

Prof. Anil Bera | Challenges and modern approach in statistics and econometrics

Prof. Anil Bera visited Spatial Warsaw for 3 weeks in June 2023. He conducted an intriguing course on Spatial Econometrics fundamentals, and provided an inspiring lecture on the topic “How to Assimilate Math, Stat and Econ for Econometrics Innovation: A Case Study of the First Nonlinear Model in Econometrics”.