"The Relationship Between Neighborhood Design and Social Capital as Measured by Carpooling," Journal of Regional Science. 2019; 59: 962-987.
Abstract: This paper examines whether social capital in a neighborhood is influenced by its design, taking the cul‐de‐sac as a special case. It offers two contributions: using carpooling to school as a proxy for social capital and a precise definition of the neighborhood using geo‐coded survey data from California. Living in a cul‐de‐sac neighborhood is found to increase the probability of carpooling, but the effect loses statistical significance when residential self‐selection is accounted for. The analysis reveals competing insights about the self‐selection behavior that must be resolved before neighborhood design can be considered an effective policy tool for enhancing social capital.
"Modeling Panel Count Data with Dynamics and Spatial Correlation"
Abstract: A Poisson-lognormal model with spatially-correlated random effects is presented for fitting panel count data of contiguous regions. The model preserves the discrete nature of the count data, incorporates dynamics in the form of separably additive lags or cumulative sum of lags, and accounts for within- and between-unit heterogeneity. An efficient Bayesian MCMC simulation algorithm is designed to estimate the model parameters. The model is used to fit data on solar panel adoption from California, and the results show the presence of spatial correlation that would otherwise lead to over-confidence in coefficient estimates if ignored. The results raise an interesting question that has been overlooked in the empirical literature about the proper specification of the dynamics underlying the spatiotemporal patterns of adoptions.
"Peer Effects and Spatial Correlation in Solar Panel Adoption"
Abstract: Motivated by the well-established result that a market failure in the form of imperfect information causes consumers to under-invest in energy-efficient technologies, this paper employed spatial econometric methods to model neighborhood peer effects and spatial correlation in the the observed spatial pattern of solar panel adoption. It contributes to the literature on peer effects in technology adoption and, hence, whether consumers’ social networks can be leveraged by policy makers to fill the information gap and accelerate adoption. Estimating a spatial transition model using individual-level data on solar panel adoption in California and Bayesian MCMC methods, the analysis finds a positive but not statistically important peer effect. The results, though, reveal that failure to control for spatially correlated unobservables leads to biased estimates.
Work in Progress
“The Governance of Spatial Change: Shaping Urban Policies and Investments in Kuwait,” with Nuno F. da Cruz.
"The effect of land use regulations on land values in Kuwait"