On the Specification of the Gravity Model of Trade: Zeros, Excess Zeros and Zero-inflated Estimation

Abstract

Conventional studies of bilateral trade patterns specify a log-normal gravity equation for empirical estimation. However, the log-normal gravity equation suffers from three problems: the

Revealed competition for greenfield investments between European regions

“…A similar taxonomy was for the different functions (see also Castellani and Pieri, 2010 Approximately one third of these investments are made by the top 500 firms. For 26,995 (98.0%) of these investments, detailed information was available regarding the NUTS-2 region in which the investment was made. Figures 1-3 show the distribution of these investments in Europe across broad sectors, economic functions, and world region of origin (see Appendix A).…”

Comparing alternative methods to estimate gravity models of bilateral trade

“…Helpman et al (2008) propose a theoretical foundation based on a model with heterogeneity of firms à la Melitz (2003) and an adapted Heckman procedure to predict trade taking into account these features. Recently, the works of Burger et al (2009), Martin and Pham (2008), Martínez-Zarzoso et al (2007), Siliverstovs and Schumacher (2009) and Westerlund and Wilhelmsson (2009) have obtained divergent results when comparing alternative estimation methods. This paper reviews most estimation methods and problems and provides a survey of the literature related to this topic.…”

Managing Fleet Capacity Effectively Under Second-Hand Market Redistribution

“…Future research in relation to this study would require the design of a new econometric specification to reduce the overestimation of the zero trade flows (e.g., a panel zeroinflated model or panel negative binomial model in line with Burger et al 2009). A dynamic panel could also be tried as an alternative to the random-effect model.…”

“…This is because of the underlying structure of our data (high number of zero flows) as discussed by Burger et al (2009). This suggests that estimates of individual trade flows should be taken with caution especially for non-zero trade flows.…”

“…Our analysis consequently follows the approach taken by Santos Silva and Tenreyro (2006) who recommend a Poisson pseudo maximum likelihood estimation. The number of transactions X ijt is assumed to follow a Poisson distribution with a conditional mean function of a set of independent variables (see Burger et al 2009 for a detailed presentation of the model). This Poisson specification allows to estimate the gravity model in its multiplicative form with trade flows entered in levels.…”