Publications

Fundamental to regional science is the subject of spatial interaction. GeoComputation - a new research paradigm that represents the convergence of the disciplines of computer science, geographic information science, mathematics and statistics - has brought many scholars back to spatial interaction modeling. Neural spatial interaction modeling represents a clear break with traditional methods used for explicating spatial interaction. Neural spatial interaction models are termed neural in the sense that they are based on neurocomputing. They are clearly related to conventional unconstrained spatial interaction models of the gravity type, and under commonly met conditions they can be understood as a special class of general feedforward neural network models with a single hidden layer and sigmoidal transfer functions (Fischer 1998). These models have been used to model journey-to-work flows and telecommunications traffic (Fischer and Gopal 1994, Openshaw 1993). They appear to provide superior levels of performance when compared with unconstrained conventional models. In many practical situations, however, we have - in addition to the spatial interaction data itself - some information about various accounting constraints on the predicted flows. In principle, there are two ways to incorporate accounting constraints in neural spatial interaction modeling. The required constraint properties can be built into the post-processing stage, or they can be built directly into the model structure. While the first way is relatively straightforward, it suffers from the disadvantage of being inefficient. It will also result in a model which does not inherently respect the constraints. Thus we follow the second way. In this paper we present a novel class of neural spatial interaction models that incorporate origin-specific constraints into the model structure using product units rather than summation units at the hidden layer and softmax output units at the output layer. Product unit neural networks are powerful because of their ability to handle higher order combinations of inputs. But parameter estimation by standard techniques such as the gradient descent technique may be difficult. The performance of this novel class of spatial interaction models will be demonstrated by using the Austrian interregional traffic data and the conventional singly constrained spatial interaction model of the gravity type as benchmark. References Fischer M M (1998) Computational neural networks: A new paradigm for spatial analysis Environment and Planning A 30 (10): 1873-1891 Fischer M M, Gopal S (1994) Artificial neural networks: A new approach to modelling interregional telecommunciation flows, Journal of Regional Science 34(4): 503-527 Openshaw S (1993) Modelling spatial interaction using a neural net. In Fischer MM, Nijkamp P (eds) Geographical information systems, spatial modelling, and policy evaluation, pp. 147-164. Springer, Berlin

Choose from multiple link options via Crossref

A character analysis reveals a clearly intermediate position of the tetraploidV. persica (2n = 28) between the two diploid speciesV. polita andV. ceratocar

During the last thirty years there has been much research effort in regional science devoted to modelling interactions over geographic space. Theoretical approaches for studying these phenomena have been modified considerably. This paper suggests a 'new modelling approach, based upon a general nested sigmoid neural network model. Its feasibility is illustrated in the context of modelling interregional telecommunication traffic in Austria and its performance is evaluated in comparison with the classical regression approach of the gravity type. The application of this neural network approach may be viewed as a three-stage process. The first stage refers to the identification of an appropriate network from the family of two-layered feedforward networks with 3 input nodes, one layer of (sigmoidal) intermediate nodes and one (sigmoidal) output node (logistic activation function). There is no general procedure to address this problem. We solved this issue experimentally. The input-output dimensions have been chosen in order to make the comparison with the gravity model as close as possible. The second stage involves the estimation of the network parameters of the selected neural network model. This is perlormed via the adaptive setting of the network parameters (training, estimation) by means of the application of a least mean squared error goal and the error back propagating technique, a recursive learning procedure using a gradient search to minimize the error goal. Particular emphasis is laid on the sensitivity of the network perlormance to the choice of the initial network parameters as well as on the problem of overlitting. The final stage of applying the neural network approach refers to the testing of the interregional teletraffic flows predicted. Prediction quality is analysed by means of two perlormance measures, average relative variance and the coefficient of determination, as well as by the use of residual analysis. The analysis shows that the neural network model approach outperlorms the classical regression approach to modelling telecommunication traffic in Austria. (authors' abstract)

Namensänderungen haben zwei ganz verschiedene Ursachen: taxonomische oder nomenklatorische. Taxonomische Änderungen gab es immer und muss es geben, solange die Wissenschaft lebendig ist. Die Nutznießer der Taxonomie kümmern sich meist nicht um diese Wissenschaft, kennen sie oft nicht, verwechseln sie mit Nomenklatur, so wie nicht selten die Taxa mit deren Namen verwechselt werden. Die Frage nach dem „richtigen Namen“ einer Pflanzensippe ist sinnlos ohne Bezug auf eine bestimmte Taxonomie, ohne Verständnis der Prinzipien von Taxonomie und Nomenklatur. Listen von Pflanzennamen können deshalb sehr missverständlich sein. Die abschließende Liste taxonomisch bedingter Namensänderungen im Wandel der Zeit führt den relativen Wert der – isoliert betrachteten – Pflanzennamen vor Augen.

The role of knowledge exchange and research cooperation between public research and the enterprise sector has received increasing attention in the analysis of innovation and technological change. The paper attempts to measure the sectoral pattern for different types of knowledge interactions and to explore the determinants of knowledge interaction between different fields of research and sectors of economic activity in Austria. The analysis is based on a comprehensive dataset on various types of knowledge interactions between university departments and private firms in Austria in the 1990s. A methodology for interaction models is used in order to identify determinants of knowledge interactions. The empirical results indicate that the intensity of knowledge interactions does not follow a simple sectoral pattern (assuming intense interactions between high-tech industries and firm-orientated technical sciences and low interactions in humanities and low-tech industries). They are rather influenced by a large set of different factors producing a complex pattern of interactions.

In the recent past, interest of Science, Technology, and Innovation (STI) policies to influence the innovation behaviour of firms has been increased considerably. This gives rise to the notion of behavioural additionality, broadening traditional evaluation concepts of input and output additionality. Though there is empirical work measuring behavioural additionalities, we know little about what role distinct firm characteristics play for their occurrence. The objective is to estimate how distinct firm characteristics influence the realisation of behavioural additionalities. We use survey data on 155 firms, considering the behavioural additionalities stimulated by the Austrian R&D funding scheme in the field of intelligent transport systems in 2006. We focus on three different forms of behavioural additionality project additionality, scale additionality and cooperation additionality and employ binary regression models to address this question. Results indicate that R&D related firm characteristics significantly affect the realisation of behavioural additionality. Firms with a high level of R&D resources are less likely to substantiate behavioural additionalities, while small, young and technologically specialised firms more likely realise behavioural additionalities. From a policy perspective, this indicates that direct R&D promotion of firms with high R&D resources may be misallocated, while attention of public support should be shifted to smaller, technologically specialised firms with lower R&D experience.

This paper provides a comprehensive review of the techniques available for categorical data and discrete choice analysis in a spatial context. The first half of the paper reviews the class of models known as generalized linear models including log-linear models,...

This paper considers the most important aspects of model uncertainty for spatial regression models, namely the appropriate spatial weight matrix to be employed and the appropriate explanatory variables. We focus on the spatial Durbin model (SDM) specification in this study that nests most models used in the regional growth literature, and develop a simple Bayesian model averaging approach that provides a unified and formal treatment of these aspects of model uncertainty for SDM growth models. The approach expands on the work by LeSage and Fischer (2008) by reducing the computational costs through the use of Bayesian information criterion model weights and a matrix exponential specification of the SDM model. The spatial Durbin matrix exponential model has theoretical and computational advantages over the spatial autoregressive specification due to the ease of inversion, differentiation and integration of the matrix exponential. In particular, the matrix exponential has a simple matrix determinant which vanishes for the case of a spatial weight matrix with a trace of zero (LeSage and Pace 2007). This allows for a larger domain of spatial growth regression models to be analysed with this approach, including models based on different classes of spatial weight matrices. The working of the approach is illustrated for the case of 32 potential determinants and three classes of spatial weight matrices (contiguity-based, k-nearest neighbor and distance-based spatial weight matrices), using a dataset of income per capita growth for 273 European regions.

This chapter describes the definitions of adverse reactions/events, pharmacovigilance systems, their limitations and basic problems, as well as adverse events connected with radiopharmaceutical agents.

Whether income levels of poorer regions are converging to those of richer is a question of paramount importance for human welfare (Islam 2003). In Europe interest in this question has been enhanced in recent years, with the entry of new countries to the European...

No abstract is provided for this article.

No abstract is provided for this article.

In several European countries there is currently a great political debate about 1993, and the structural and economic changes which will have to come about in Europe within the coming years. Substantial industrial restructuring is taking place and forward planning in...

This paper lies in the tradition of decomposition analysis of wage differentials based on the model set forth in Blinder (1973) and Oaxaca (1973), and aims to measure labour market discrimination against Roma in South East European countries (Albania, Bulgaria, Croatia, Serbia and Kosovo). We use microdata from 2004 UNDP household survey and a Bayesian approach, proposed by Keith and LeSage (2003), for the decomposition analysis of wage differentials. Statistical inference for both discrimination and characteristics effects estimates are based on Markov Chain Monte Carlo (MCMC) estimation. Variance estimates derived from this method of estimation are known to reflect the true posterior variance when a sufficiently large sample of MCMC draws is carried out. The results provide clear evidence for labour market discrimination against Roma in Albania and Kosovo, but not so in Bulgaria, Croatia, and Serbia. Nevertheless, there are significant differences in how individual characteristics are valued between Roma and non-Roma.