This paper compares the performance of Bayesian variable selection approaches for spatial autoregressive models. We present two alternative approaches which can be implemented using Gibbs sampling methods in a straightforward way and allow us to deal with the problem of model uncertainty in spatial autoregressive models in a flexible and computationally efficient way. In a simulation study we show that the variable selection approaches tend to outperform existing Bayesian model averaging techniques both in terms of in-sample predictive performance and computational efficiency. (authors' abstract) / Series: Department of Economics Working Paper Series

Submitted by Israel Vieira Neto (israel.vieiraneto@ufpe.br) on 2015-03-06T14:22:22Z No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) DISSERTAÇÃO GIUSEPPE TREVISAN.pdf: 3587579 bytes, checksum: fa47c846ce99688bf17f94c5df29eb87 (MD5) / Made available in DSpace on 2015-03-06T14:22:22Z (GMT). No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) DISSERTAÇÃO GIUSEPPE TREVISAN.pdf: 3587579 bytes, checksum: fa47c846ce99688bf17f94c5df29eb87 (MD5) Previous issue date: 2013-03-08 / FACEPE / Um segmento importante da literatura de Economia do Crime afirma que, além das variáveis socioeconômicas, o espaço é fator fundamental a se associar com a criminalidade. Dada a falta de evidências empíricas sobre a relação entre espaço e crime, este estudo tem por objetivo identificar as correlações entre as variáveis ambientais com a taxa de homicídio nos bairros da cidade do Recife. Para isso, foram construídas variáveis de ambiente que capturam características relacionadas à distribuição dos tipos de domicílios dos bairros do Recife e foi implementada a técnica de econometria espacial para averiguar efeitos de spillover espaciais. O modelo SAR apresenta o melhor ajuste e mostra que a proporção de estabelecimentos nãoresidenciais em relação ao total de estabelecimentos do bairro tem uma relação positiva com a criminalidade e a concentração de domicílios residenciais está associada a índices de criminalidades mais baixos. As correlações das variáveis socioeconômicas seguem o padrão da maioria dos achados da literatura nacional e internacional, exceto para o caso da densidade demográfica.

<div> <div> <div> <p>First Chapter: The ordinary least squares (OLS) estimator for spatial autoregressions may be consistent as pointed out by Lee (2002), provided that each spatial unit is influenced aggregately by a significant portion of the total units. This paper presents a unified asymptotic distribution result of the properly recentered OLS estimator and proposes a new estimator that is based on the indirect inference (II) procedure. The resulting estimator can always be used regardless of the degree of aggregate influence on each spatial unit from other units and is consistent and asymptotically normal. The new estimator does not rely on distributional assumptions and is robust to unknown heteroscedasticity. Its good finite-sample performance, in comparison with existing estimators that are also robust to heteroscedasticity, is demonstrated by a Monte Carlo study.<br></p><p><br></p><p>Second Chapter: This paper proposes a new estimation procedure for the first-order spatial autoregressive (SAR) model, where the disturbance term also follows a first-order autoregression and its innovations may be heteroscedastic. The estimation procedure is based on the principle of indirect inference that matches the ordinary least squares estimator of the two SAR coefficients (one in the outcome equation and the other in the disturbance equation) with its approximate analytical expectation. The resulting estimator is shown to be consistent, asymptotically normal and robust to unknown heteroscedasticity. Monte Carlo experiments are provided to show its finite-sample performance in comparison with existing estimators that are based on the generalized method of moments. The new estimation procedure is applied to empirical studies on teenage pregnancy rates and Airbnb accommodation prices.<br></p><p><br></p><p>Third Chapter: This paper presents a sample selection model with spatial autoregressive interactions and studies the maximum likelihood (ML) approach to estimating this model. Consistency and asymptotic normality of the ML estimator are established by the spatial near-epoch dependent (NED) properties of the selection and outcome variables. Monte Carlo simulations, based on the characteristics of female labor supply example, show that the proposed estimator has good finite-sample performance. The new model is applied to empirical study on examining the impact of climate change on agriculture in Southeast Asia.<br></p></div></div></div><div><div><div> </div> </div> </div>

This paper investigates the nexus between regional income levels and inequality. We present a novel small-scale inequality database for Austrian municipalities to address this question. Our dataset combines individual tax data of Austrian wage tax payer on regionally disaggregated scale with census and geographical information. This setting allows us to investigate regional spillover effects of average income and various measures of income inequality. Using this data set we find distinct regional clusters of both high average wages and high earnings inequality in Austria. Furthermore we use spatial econometric regressions to quantify the effects between income levels and a number of inequality measures such as the Gini and 90/10 quantile ratios. (authors' abstract) / Series: Department of Economics Working Paper Series

In this study, we extend the spatial weight matrix defined by Getis and Aldstadt (2004) to a more general case. The modified spatial weight matrix performs better than the original spatial weight matrix since the modified spatial weight matrix adjusts weights of observations based on the distance from other observations. Both the simulation study and the application to the ecological process of invasion of non-native invasive plants (NNIPs) provide evidences for the better performance of the modified spatial weight matrix. We also develop procedures that can be used to quantify the invasion stages of NNIPs. The resultant map of invasion stage on county-level provides a useful and meaningful tool for policy makers; especially, it can be used to optimize allocation of management resources. The result of simultaneous autoregressive model shows that not only the biotic and abiotic factors but also human activities play an important role in the establishment and spread of multiflora rose in the Upper Midwest. It also shows the tendency of the establishment and spread of multiflora rose (Rosa Multiflora, Thunb. ex Murr.) in the Upper Midwest.

No description available.

This study focuses on how spatial circumstances affect property owners’ preference toward sustainable urban development, in the form of three-essays. In the first essay, property owners’ preference toward the concept of compact development is identified. Compact development is an increasingly popular concept that includes multiple aspects, such as mixed land use, high density, and pedestrian/transit-friendly options. Previous hedonic literature on the comprehensive effect of compact development is limited. Also, spatial dependence in the data, something likely endemic to compact development, has not yet been thoroughly addressed. This study uses a spatial fixed-effect model, a spatial-autoregressive model with auto-regressive disturbances (SARAR), and a spatial fixed-effect SARAR model to determine the price effect of “compactness” in a major U.S. metropolitan area. By analyzing of 47,000 sales records in Fulton County over a decade, this study indicates that home buyers prefer to have smaller, more diffuse greenspace nearby, rather than a large, concentrated greenspace at a longer walking distance. High parcel density and diverse land use is consistently disvalued, and the premium on accessing public transportation is not identified among all models. No specific trend over time has been observed, despite the recession starting in 2008. Finally, a comprehensive index of compactness shows relatively high willingness-to-pay for compact development. The second essay tests the spatial spillover of signaling within the pursuit of LEED certification. The benefit of pursuing green building certification mainly comes from two aspects: the cost-effectiveness from energy efficiency and the signaling consideration, including the premium on property values, benefits from a better reputation, morality values, or purely pride. By analyzing all new constructions that received LEED certification from 2000 to 2012 (LEED-NC v2.0 to v2.2) in the U.S., this study tries to identify the size of the signaling effects, and spillover of signaling, as building owners pursue LEED certification. The results show that the signaling effect affects decision making in pursuing LEED certification, especially at scores around thresholds. The size of signaling effects differs among different owner types and different certificate levels. For the Gold level or below, government and non-profit-organization owners value signaling more than do profit-seeking firms. At the Platinum level, there is no significant difference among owner types. This study also finds that the signaling effect clusters spatially for government and profit-seeking firms. Finally, the results show that the cluster of signaling is independent from the cluster of LEED buildings, indicating that mechanisms behind the cluster of signaling are different from those of LEED constructions. The third essay tests the distance effect on the support for Atlanta BeltLine. Atlanta BeltLine, a large urban redevelopment project currently underway in the center of Atlanta, transforms 22 miles of historical railroad corridors into parks, trails, pedestrian-friendly transit areas, and affordable housing. This study aims to determine the distance effect on the support of Atlanta BeltLine and whether the implement of Tax Increment Financing (TIF) affects the support. The contributions of this exercise are twofold. First, it demonstrates the risks and remedies to missing spatial data by solving the technical problem of missing precise spatial location values. Second, it tests underlying reasons why distance can help explain the level of support that Atlanta BeltLine has received, with striking implications for theories like the Homevoter hypothesis. Survey data used in this study was conducted in summer 2009, about three years after the declaration of the project. The support by both homeowners and renters significantly declines as distance from the BeltLine increases. However, when residents’ tendency to use BeltLine parks and transits is entered as a variable, the distance effect disappears. By indicating that the distance effect comes from homeowners’ and renters’ the accessibility to BeltLine amenities, the result rejects the homevoter hypothesis, which holds that property value increment is the main mechanism behind support. The results also show that whether or not a homeowner or renter is a parent in City of Atlanta affects a person’s support of the BeltLine. These results lead to the conclusion that the concern of TIF affecting future school quality hampers the support of the project.

No description available.