Lw Hepple

Bayesian Techniques in Spatial and Network Econometrics: 2. Computational Methods and Algorithms

Bayesian theory has been seen as having considerable potential and attractiveness for model estimation and analysis in spatial and network econometrics. However, analytical and computational problems have also been seen as a great barrier. In this paper the analytical simplifications available are developed and the algorithms required are examined. The author argues that, for a broad class of models in spatial econometrics, Bayesian analysis is quite practicable and can be implemented without great cost. The spatial specifications are mapped into the various forms of Bayesian computation available and detailed examples are provided. Recent developments on the frontier of Bayesian computation have potential to expand further the practical applicability of the Bayesian approach to spatial econometrics.

Bayesian Techniques in Spatial and Network Econometrics: 1. Model Comparison and Posterior Odds

In this paper the problems of specification and nonnested model comparison in spatial and network econometrics are examined, and the Bayesian posterior probabilities approach is developed. The theory is developed for the comparison of alternative spatial weights matrices in both the systematic and the disturbance components of models, and also for the comparison of alternative spatial disturbance processes. Several empirical illustrations are provided, and extensions of the Bayesian approach are discussed.

The revival of geopolitics

Abstract This paper chronicles the revival of geopolitical writing and analysis in North America and Europe since 1970, after 20 years of decline. The revival is examined in terms of both language and substance, and the reasons for the revival are explored. As well as its role in the analysis of global and regional strategy, geopolitics has entered popular language and political discourse. The contributions of geographers are discussed, and the opportunities and dangers of the revival are examined. The importance of historical and political critique is argued.

Exact testing for spatial autocorrelation among regression residuals

This paper develops the exact distribution of the Moran or Cliff-Ord test for spatial autocorrelation among regression residuals. It shows how practical, computational methods can be used to implement the exact test. Higher order approximations to the distribution are also discussed. Empirical applications are provided, and the value of exact testing is discussed.

Bayesian analysis of the linear model with spatial dependence

This paper develops a Bayesian analysis of the linear regression model with spatial dependence in the disturbances. The existing literature in spatial econometrics has been based entirely on sampling theory approaches (e. g. Neyman-Pearson and maximum likelihood) to statistical inference, and the Bayesian perspective has not been explored for spatial estimation. In contrast, the mainstream of econometric work has been influenced by the Bayesian approach to statistical inference during the last ten years, largely through the texts and papers of Zellner and Box (Zellner, 1971; Box and Tiao, 1973) and their associates and students. Bayesian methods have been fruitfully applied to the analysis and estimation of a wide range of econometric problems, such as simultaneous equations (Chetty, 1968), production functions (Tsurumi and Tsurumi, 1976; Zellner and Richard, 1973), distributed lag models (Zellner and Geisel, 1970), the linear model with serially correlated errors (Zellner and Tiao, 1965), and the linear model with non-normal errors (Box and Tiao, 1962).

Geopolitics, generals and the state in Brazil

Abstract This study traces the history of geopolitics in Brazil and the major contributions made by army generals, notably General Golbery do Couto e Silva. The relationship that developed in the 1950s between geopolitics, national security doctrine and military ideology is then examined. The paper argues that geopolitics was an important source of a theory of the state for the military regime, but that the specific role of geopolitics can be overstated. This analysis is then related to wider debates about military-authoritarian regimes in South America

BAYESIAN MODEL CHOICE IN SPATIAL ECONOMETRICS

Within spatial econometrics a whole family of different spatial specifications has been developed, with associated estimators and tests. This lead to issues of model comparison and model choice, measuring the relative merits of alternative specifications and then using appropriate criteria to choose the “best” model or relative model probabilities. Bayesian theory provides a comprehensive and coherent framework for such model choice, including both nested and non-nested models within the choice set. The paper reviews the potential application of this Bayesian theory to spatial econometric models, examining the conditions and assumptions under which application is possible. Problems of prior distributions are outlined, and Bayes factors and marginal likelihoods are derived for a particular subset of spatial econometric specifications. These are then applied to two well-known spatial data-sets to illustrate the methods. Future possibilities, and comparisons with other approaches to both Bayesian and non-Bayesian model choice are discussed.

Context, Social Construction, and Statistics: Regression, Social Science, and Human Geography

In response to a paper by T J Barnes, published in 1998, the author accepts the same social-constructivist perspective, but argues that the structure of regression was not excessively constrained by its biometric origins. The history of regression and its use in the social sciences is examined, and the author argues that any assessment of regression in human geography must be set against this wider context.

Multiple Regression and Spatial Policy Analysis: George Udny Yule and the Origins of Statistical Social Science

Studies on the history of statistics by MacKenzie and on quantitative geography by Barnes have suggested that the lineaments and assumptions of statistical methods such as correlation and regression are closely related to their origin in biometrics and eugenics. This paper challenges that view by examining in detail the work of George Udny Yule. Yule was a colleague of Karl Pearson in the 1890s, but was interested in social science and social policy applications, not eugenics. In the late 1890s he constructed both the theory and application of multiple regression analysis, using geographical data. The paper examines Yules work and its context, relating it to debates on the history of statistics, and traces the subsequent early diffusion of regression and correlation into the social sciences. The paper concludes by arguing for greater recognition of Yules pivotal role, and also for further studies on the history of quantitative social science.

Socialist Geography in England: J. F. Horrabin and a Workers' Economic and Political Geography

This paper examines the life and geographical writings of the English socialist J.F.(“Frank”) Horrabin (1884–1962) and his attempts to construct a socialist geography. Horrabin was an active socialist in the Labour Party, Fabian Society, and other leftwing groups and very involved in working-class education through the Plebs League and National Council of Labour Colleges. He was also a journalist, cartoonist, and gifted cartographer. His 1923 text An Outline of Economic Geography, which sold in large numbers and was translated into nine other languages, attempted to provide workers with an account of economic (and political and historical) geography that used bourgeois “pure geography” but put it within a socialist and historical–materialist framework. Unlike Germany and some other countries, England did not have a strong Marxist theoretical tradition, and Horrabins approach does not develop theory (though it did attract the admiration of the German Marxist Karl Wittfogel). Rather, it sets out to be engaged in practical political education. Horrabins work was developed within a particular context, but his geographical writings (and pioneering political cartography) exemplify one way of linking geography with political practice, and this paper examines these relationships. Many of Horrabins concerns find echoes in current radical geography, and his work deserves belated recognition and a place in the history of geography.