Replication Files

Postdoctoral Research Proposals 

1. Postdoctoral in Econometrics: 

During September 2023, I joined the Faculty of Social Sciences at the University of Helsinki as a Postdoctoral Researcher for the research project: "Structural Analysis of Non-Gaussian Time Series with Applications to Climate and Macroeconomy" (funded by the Research Council of Finland). 

2. Independent Project Research Proposal: 

"A Spatio-Temporal Autoregressive Distributed Lag Model with an Application to Monitoring Spatial Heterogeneity in Ecological Dynamics of Marine Habitats" (January 2024).

3. Postdoctoral Research Applications for/during the academic year 2024/2025.

Doctoral Research Proposals

• PhD Research Proposal (Department of Economics, University of Southampton, February 2018). 

"Essays on the Econometrics of Financial Networks: Theory and Applications".

• PhD Student Application Proposal (submitted at the University of Southampton, July 2017). 

"Essays on Cointegration Dynamics and Market Integration: Modelling Cross-Country Convergence Dynamics".

Submitted PhD research proposal for consideration to a University of Southampton PhD Scholarship  under the supervision of Prof. Jose Olmo and Prof. Jean-Yves Pitarakis (accepted).

• PhD Student Application Proposal (submitted at the University of Warwick Business School, April 2017). 

"Essays on Economic Modelling and Forecasting: Forecasting Time Series under Structural Breaks".

 Submitted PhD research proposal for consideration to an Office of National Statistics (ONS) funded research project offering a PhD Scholarship at WBS under the supervision of Prof. James Mitchell and Associate Prof. Ana Galvao (rejected). 

The Econometrics of Networks: Theory and Applications  

Summary:

Understanding the dynamics of socioeconomic and financial networks is crucial in examining related transmission processes such as the amplification of financial crises through systemic risk, the efficiency of public investments across local governments, the diffusion of entrepreneurial finance into currently unexplored industries, the polarization of political opinions as well as the dynamics of cross-country intergenerational mobility. Due to the interconnectedness of economic agents (e.g., households, firms, economies) and the dynamic structure of the network under specific conditions these phenomena exhibit persistence behaviour while coexistence of threshold effects produce multiple equilibrium. Furthermore, the high degree of interconnectedness between financial markets and embeddedness of firms within and across economies urge econometricians to provide robust tools for accurately examining the structure, mechanisms and implications of economic networks such as the spillover and network effects of local to neighboring and global economies. 

The current literature on modelling methodologies for network structures spans the Social Interactions Models of Manski, the Durbin Spatial Autoregressive Models as well as the Friedkin-Johnsen Model of spatial effects among others. With the proposed research study, we aim to provide a unified framework to examine the theoretical aspects of the econometrics of networks, including model identification and statistical computation techniques, model validation techniques e.g., measurement error and model misspecification, especially in cases of endogenous networks which might exhibit switching or dynamic features. Moreover, using state-of-the-art econometric methodologies we aim to decompose the sources of heterogeneity and possible parameter fragility across different levels of aggregation. More specifically, emphasis will be given in examining certain econometric features useful in dynamic network frameworks such as various network centrality and connectedness measures and statistical tests (e.g. CoVaR, Network Granger causality tests), the properties of adjacency matrices which can model the dynamic behaviour of economic agents within networks, as well as economic aspects which can affect the amplification of network effects (e.g., risk aversion, credit risk, business cycle risk etc.). Robustness evidence will be provided through simulated studies and empirical applications of related financial and macroeconomic aspects. 

The economic applications of the proposed research project include the study of spillover effects from R&D and entrepreneurial investments within and across economies, the identification of peer effects in financial networks and the credit provision or SMEs, as well as the transmission of systemic risk and financial contagion via financial networks. For example, certain spatially correlated variables and high frequency estimators can be used as proxies for unobserved drivers of heterogeneity and market volatility. Lastly, the proposed methodological approach will allow us to better examine the micro-foundation properties of economic behaviour which diffuses in financial markets and consequently to the real economy. In particular, according to Glasserman and Young (2016) there is “a clear trade-off between the stabilizing effect of interconnections due to diversification, and the amplifying effect from additional channels through which shocks can spread”. The long term benefits of the proposed research study is the acquisition of knowledge in modelling network dynamics across different economic phenomena. Therefore, using linear and nonlinear spatial models will allow us to study the dynamic behaviour of economic agents such as spillover effects and peer effects, which occur due to an underline network structure. Think for example, innovation clusters, consumption patterns or financial networks; robust econometric methods can accurately estimate these unobserved individual heterogeneous effects, which is of paramount importance for informative policy making.

PhD Research Proposal Examiners: 

• Dr. Maria Kyriakou (Research Field: Spatial Econometrics) 

• Prof. Jean-Yves Pitarakis (Research Field: Econometrics).

References:

• Shi, W., and Lee, L. F. (2018). "A Spatial Panel Data Model with Time Varying Endogenous Weights Matrices and Common Factors". Regional Science and Urban Economics, 72, 6-34. 

• Qu, X., Lee, L. F., and Yu, J. (2017). "QML Estimation of Spatial Dynamic Panel Data Models with Endogenous Time Varying Spatial Weights Matrices". Journal of Econometrics, 197(2), 173-201. 

• Zhu, X., Pan, R., Li, G., Liu, Y., and Wang, H. (2017). "Network Vector Autoregression". Annals of Statistics, 45(3): 1096-1123. 

• Barigozzi, M., and Hallin, M. (2017). "A Network Analysis of the Volatility of High Dimensional Financial Series". Journal of the Royal Statistical Society: Series C (Applied Statistics), 66(3), 581-605.

• Blasques, F., Koopman, S. J., Lucas, A., and Schaumburg, J. (2016). "Spillover Dynamics for Systemic Risk Measurement using Spatial Financial Time Series Models". Journal of Econometrics, 195(2), 211-223. 

• Gilchrist, D. S. (2016). "Patents as a Spur to Subsequent Innovation? Evidence from Pharmaceuticals". American Economic Journal: Applied Economics, 8(4), 189-221. 

• Glasserman, P., and Young, H. P. (2016). "Contagion in Financial Networks". Journal of Economic Literature, 54(3), 779-831.

• Glasserman, P., and Young, H. P. (2015). "How likely is Contagion in Financial Networks?". Journal of Banking & Finance, 50, 383-399.

• Leary, M. T., and Roberts, M. R. (2014). "Do Peer Firms affect Corporate Financial Policy?". Journal of Finance, 69(1), 139-178. 

• Acemoglu, D., Carvalho, V. M., Ozdaglar, A., and Tahbaz‐Salehi, A. (2012). "The Network Origins of Aggregate Fluctuations". Econometrica, 80(5), 1977-2016.

• Yu, J., de Jong, R., and Lee, L. F. (2012). "Estimation for Spatial Dynamic Panel Data with Fixed Effects: The Case of Spatial Cointegration". Journal of Econometrics, 167(1), 16-37. 

• Yu, J., De Jong, R., and Lee, L. F. (2008). "Quasi-maximum Likelihood Estimators for Spatial Dynamic Panel Data with Fixed Effects when both n and T are large". Journal of Econometrics, 146(1), 118-134. 

• Bai, J. (2009). "Panel Data Models with Interactive Fixed Effects". Econometrica, 77(4), 1229-1279. 

Remark 1: Econometric specifications which capture network dependence aim to model economic behaviour which is often dynamic, that is, it is influenced by past own behaviour. Without loss of generality, dynamic econometric models permit to capture "state dependence" which appears due to habit formation, market and social interactions among other. Recently, it has been pointed out that an economic agent's own behaviour is also influenced by the behaviour of other agents, typically their peers. This is due to network linkages, social interactions and spillover effects (e.g., see Cui et al (2022, Econometric Journal)).

Remark 2: The estimation and inference approach proposed by Ando, Li and Lu (JoE, 2023) corresponds to conditional quantile specification which allows to jointly model spatial interactions and common shocks for a fixed quantile level with respect to spatial spillover effects. Their framework treats the spatial matrix as exogenous (i.e., non-stochastic weight matrix). Moreover, an algorithmic procedure (see Algorithm 1) is proposed to obtain estimates for factor loadings and other unknown model parameters. Allowing for the presence of nonstationary factors is more challenging; especially since spatial-based specifications require the development of novel asymptotic theory.  

Ex-Post Related Literature:

• Chen, J., Cui, G., Sarafidis, V. and Yamagata, T. (2025). "IV Estimation of Heterogeneous Spatial Dynamic Panel Models with Interactive Effects". Preprint arXiv:2501.18467.

• Wang, W., Wooldridge, J. M., Xu, M., Lu, C., and Zheng, C. (2024). "Using Generalized Estimating Equations to Estimate Nonlinear Models with Spatial Data". Econometric Reviews, 1-29. 

• Ando, T., Li, K., and Lu, L. (2023). "A Spatial Panel Quantile Model with Unobserved Heterogeneity". Journal of Econometrics, 232(1), 191-213. 

• Jin, F., and Wang, Y. (2023). "Consistent Non-Gaussian Pseudo Maximum Likelihood Estimators of Spatial Autoregressive Models". Econometric Theory, 1-39. 

• Cui, G., Sarafidis, V., and Yamagata, T. (2023). "IV estimation of Spatial Dynamic Panels with Interactive Effects: Large Sample Theory and an Application on Bank Attitude Towards Risk". The Econometrics Journal, 26(2), 124-146.

• Olmo, J., and Sanso-Navarro, M. (2023). "A Nonparametric Spatial Regression Model using Partitioning Estimators". Econometrics and Statistics. 

• Armillotta, M., and Fokianos, K. (2023). "Nonlinear Network Autoregression". Annals of Statistics, 51(6), 2526-2552. 

• Martellosio, F., and Hillier, G. (2020). "Adjusted QMLE for the Spatial Autoregressive Parameter". Journal of Econometrics, 219(2), 488-506. 

• Foglia, M., and Angelini, E. (2019). "The Time-Spatial Dimension of Eurozone Banking Systemic Risk". Risks, 7(3), 75. 

• Lu, C., Wang, W., and Wooldridge, J. M. (2018). "Using Generalized Estimating Equations to Estimate Nonlinear Models with Spatial Data". Preprint arXiv:1810.05855.