1,721,016 research outputs found
Moscone, F., Tosetti, E. (2010), Testing for Spatial Correlation in Panels with Unobserved Common Effects: an Application to US Health Expenditure
No full textGMM estimation of Spatial Panels with Fixed Effects and Unknown Heteroskedasticity
No full textIn this paper we consider the estimation of a panel data regression model with spatial autoregressive disturbances, fixed effects and unknown heteroskedasticity. Following the work by Kelejian and Prucha (1999), Lee and Liu (2006a) and others, we adopt the Generalized Method of Moments (GMM) and consider moments as a set of linear quadratic conditions in the disturbances. As in Lee and Liu (2006a), we assume that the inner matrices in the quadratic forms have zero diagonal elements to robustify moments against unknown heteroskedasticity. We derive the asymptotic distribution of the GMM estimator based on such conditions. Hence, we carry out some Monte Carlo experiments to investigate the small sample properties of GMM estimators based on various sets of moment conditions. © 2011
Robust estimation under error cross section dependence
Full text linkThis article has been made available through the Brunel Open Access Publishing Fund.We propose a robust, partial sample estimator for the covariance matrix of the fixed effects and mean
group estimators of the slope coefficients in a short T panel data model with group-specific effects and
errors that are weakly cross sectionally dependent and serially correlated
HAC estimation in spatial panels
Full text link© 2012 Elsevier B.V. All rights reservedWe propose a HAC estimator for the covariance matrix of the fixed effects estimator in a panel data model with unobserved fixed effects and errors that are both serially and spatially correlated.conomic and Social Research Council (grant RES-061-25-0317)
A review and comparison of tests of cross section independence in panels
No full textIn this paper we review and compare diagnostic tests of cross-section independence in the disturbances of panel regression models. We examine tests based on the sample pairwise correlation coefficient or on its transformations, and tests based on the theory of spacings. The ultimate goal is to shed some light on the appropriate use of existing diagnostic tests for cross-equation error correlation. Our discussion is supported by means of a set of Monte Carlo experiments and a small empirical study on health. Results show that tests based on the average of pairwise correlation coefficients work well when the alternative hypothesis is a factor model with non-zero mean loadings. Tests based on spacings are powerful in identifying various forms of strong cross-section dependence, but have low power when they are used to capture spatial correlation. © Journal compilation © 2009 Blackwell Publishing Ltd
Health expenditure and income in the United States
No full textThis paper investigates the long-run economic relationship between health care expenditure and income in the US at a State level. Using a panel of 49 US States over the period 1980-2004, we study the non-stationarity and co-integration between health spending and income, ultimately measuring income elasticity of health care. The tests we adopt allow us to explicitly control for cross-section dependence and unobserved heterogeneity. Specifically, in our regression equations we assume that the error has a multifactor structure, which may capture global shocks and local spill overs in health expenditure. Our results suggest that health care is a necessity rather than a luxury, with an elasticity much smaller than that estimated in other US studies. Further, we detect significant spatial concentration in US health spending. Our broad perspective of cross-section dependence as well as the methods used to capture it give new insights on the debate over the relationship between health spending and income. Copyright © 2009 John Wiley & Sons, Ltd
Testing for error cross section independence with an application to US health expenditure
No full textThis paper considers the problem of testing for error cross section independence in a panel where statistical units may be subject to unobserved common effects, spatial spill overs, or both. We review a number of diagnostics that are used for testing for error cross section independence in panels, including tests based on spacings and spatial statistics. We then argue that commonly used spatial statistics might give misleading results when cross section correlation arising from common effects is not taken into account. Hence, we study the properties of spatial statistics applied to residuals obtained from an augmented regression, where common factors have been approximated by principal components (Bai, 2009). Small sample properties of our testing strategy are investigated in a Monte Carlo study. Results show that spatial tests applied to de-factored residuals detect well the presence of spatial correlation in the data. The paper concludes with a small empirical exercise on US health expenditure. © 2009 Elsevier B.V
Weak and strong cross-section dependence and estimation of large panels
No full textThis paper introduces the concepts of time-specific weak and strong cross-section dependence, and investigates how these notions are related to the concepts of weak, strong and semi-strong common factors, frequently used for modelling residual cross-section correlations in panel data models. It then focuses on the problems of estimating slope coefficients in large panels, where cross-section units are subject to possibly a large number of unobserved common factors. It is established that the common correlated effects (CCE) estimator introduced by Pesaran remains asymptotically normal under certain conditions on factor loadings of an infinite factor error structure, including cases where methods relying on principal components fail. The paper concludes with a set of Monte Carlo experiments where the small sample properties of estimators based on principal components and CCE estimators are investigated and compared under various assumptions on the nature of the unobserved common effects. © 2011 The Author(s). The Econometrics Journal © 2011 Royal Economic Society
Real estate market and financial stability in US metropolitan areas: A dynamic model with spatial effects
Full text linkThis paper investigates spatio-temporal variations in ex-post credit risk in the United States, as a function of real estate prices, loan purchases made by government sponsored enterprises, and a set of local characteristics during the recent housing boom and bust.We model bank's non-performing loans as a first-order dynamic panel data regression model with group-specific effects and spatial autoregressive errors. To estimate this model, we develop an ad-hoc generalized method of moments procedure which consists of augmenting moments proposed by the panel literature to estimate short T, pure dynamic panels, with a set of quadratic conditions in the disturbances. Results on estimation of the empirical model point at the negative impact of real estate prices on non-performing loans. Further, our results show that a rise in the number of real estate mortgages backed by government-sponsored enterprises increases non-performing loans, thus deteriorating the quality of banks' loan portfolio.The ESRC (Ref. no. RES-061-25-0317)
Social Interaction in Patients'Hospital Choice: Evidence from Italy
No full text. We study the influence of social interaction on patients choice of hospital and its relationship with the quality that is delivered by hospitals, using Italian data. We explore the effect on individual choices of a set of variables such as travel distance and individual- and hospital-specific characteristics, as well as a variable capturing the effect of the neighbourhood. The richness of our data allows us to disentangle the influence of sharing information (the network) on patients choices of hospital from contextual effects. Our empirical investigation suggests that past experience in the utilization of health services by the network plays a significant role in explaining current patients choices of hospital. Other relevant factors that influence patients decisions of being admitted in a particular hospital are prior use of health services in that hospital, patient-to-hospital distance and supply factors such as the number of beds and number of doctors. We then investigate the relationship between a set of health outcome indicators and the sensitivity of patients choices to the network, to test whether sharing information increases the likelihood of selecting a high quality hospital. Our results suggest that social interaction does not have an influence on health outcomes, and in some cases it may even mislead patients, who end up in low quality institutions. One explanation for this result is the absence of a source of information on the quality of hospitals that is accessible to all individuals, such as guidelines or star ratings, which may exacerbate the influence of information that is gathered locally on choices of hospital and may result in a lower degree of competition between hospitals and lower quality
- …
