My Research

Published Article

"Experiential Learning, Sense of Belonging and Feedback using Economics Games.", joint with Antonio Rodriguez-Gil and Ali Raza, Published in The Economics Network. https://doi.org/10.53593/n4256a.

Working Papers 

"Estimation with Multiple and Different Cross-sectional Structural Breaks in Heterogeneous Spatial

Panels with Latent Multi-factors.", joint with Kausik Chaudhuri.

Abstract: This study proposes a novel econometric method for estimation with multiple structural breaks in heterogeneous panel data models, accounting for latent multiple factor structure and spatial dependence. Utilising the CCEX-2SLS technique, our approach achieves consistent and asymptotically valid individual and mean group estimators, offering significant improvements over traditional methods. A key innovation is the method’s flexibility in allowing for different cross-sectional breaks, in other words, individual-specific break points in panel data analysis, which is particularly important in contexts where shocks propagate unevenly across spatial units or with time delays. Monte Carlo simulations demonstrate the strong finite-sample performance of our proposed method. Through an empirical application on the UK local housing market—covering 345 local authority districts across regions—we demonstrate the method’s practical relevance. The analysis reveals that major economic events, such as the 2008 financial crisis and the Help to Buy (HtB) schemes, had substantial and regionally differentiated impacts on house prices. Our results highlight the importance of accounting for spatial dependence and heterogeneous structural breaks in panel data analysis. This framework provides new insights for future research into dynamic processes, with potential extensions including formal testing for common versus different cross-sectional break structures.

"Uncovering Spatial Spillovers in UK Housing Markets: A Machine Learning Network Approach.", joint with Kausik Chaudhuri.

Abstract: Prior studies examining spatial effects have predominantly relied on geographic proximity as the main driver of spatial dependence among local housing markets. However, such geographic relationships may not fully capture the complex mechanisms underlying spatial linkages between housing submarkets. Existing research has rarely analysed the determinants of the underlying spatial spillover network in housing price panels. To fill this gap, we contribute by utilising innovative methodologies that integrate Adaptive Elastic Net GMM to identify latent spatial spillover matrices for housing prices across five aggregated regions in England and Wales. These identified matrices are then used to examine the driving factors behind spillovers through dyadic regressions, employing various spatial weight matrices constructed from geographic proximity, internal migration patterns, and local economic conditions. Our determinant analysis shows that not only geographic contiguity and internal population flows, but also disparities or proximities in economic conditions, govern spatial interdependencies across local housing markets. We further validate the identified spatial matrices using Global Moran’s I tests and Variable Importance Analysis (VIP) via Random Forest and Elastic Net techniques, thereby enhancing the reliability of our findings.

"Does Homogeneity and Spatial Spillovers Challenge Market Efficiency in Local House Markets? Evidence

from UK.", joint with Kausik Chaudhuri and Sandra Lancheros.

Abstract: This paper investigates the efficiency of UK regional housing markets by evaluating the roles of temporal and spatial dynamics in price formation. We examine whether local house price movements are better explained by homogeneous national factors, heterogeneous regional conditions, or spatial interdependence across regions. Using monthly data at the Local Authority District (LAD) level, we estimate and compare the forecasting performance of three models: Heterogeneous Autoregressive (Heter-AR), Homogeneous AR (Homo-AR), and Network AR (NAR). Our findings show that regional housing markets exhibit persistent heterogeneity and significant short-term spatial spillovers, particularly in the North of England. Spatial impulse response analysis reveals that price shocks in one region can propagate to others but decay within 12 months, highlighting localized contagion. Forecast evaluation using RMSE, relative MSPE, and the Diebold-Mariano test demonstrates that the NAR model outperforms both Heter-AR and Homo-AR models, suggesting that spatial and national effects jointly enhance predictability. These results challenge the Efficient Market Hypothesis (EMH) by showing that past regional and spatial price information contains significant forecasting power. The paper contributes to debates on housing market efficiency and offers practical implications for forecasting, policy formulation, and regional housing planning.

Working in Progress

"Fiscal–Monetary Interactions in Times of Elevated Sovereign Debt: Evidence in EU, UK and USA.", joint with Muhammad Ali Nasir and Kausik Chaudhuri.

"A New Insight of Regime Shift and Inter-Group Inference with Interactive Effects.", joint with Kausik Chaudhuri.

"Inequality of Opportunity under Current China’s License Plate Policy.", joint with Nina Xiaochun Sun and Zaifang He.