An Econometric Estimation of Industrial Water Demand in France

This study investigates the structure of industrial water demand byestimating the derived demand for water on a sample of industrialestablishments located in the south-west of France. Productiontechnologies are represented by short-term variable cost functions andapproximated by a translog form. Industrial water use is modeled ashaving three components: the quantity of water bought to a waterutility, the quantity of autonomous water and the quantity of watertreated prior to use. We include in this framework water effluents,which are considered as a by-product of the production process, emittedby firms. Each of the three water components is treated as a separateinput and all are estimated as a system of simultaneous equations. Themodel is estimated on a sample of 51 industrial plants in the Girondedistrict observed from 1994 to 1996 using Seemingly Unrelated Regression(SUR) and Feasible Generalized Least Squares (FGLS). Results ofestimations show that industrial firms are sensitive to water priceinputs. Network water elasticity is estimated at –0.29. It variesfrom –0.10 to –0.79 according to the type of industryconsidered. Autonomous water price elasticity is not significant.Elasticity for treated water is evaluated at –1.42 at themean-sample and varies from –0.90 to –2.21 according to theindustry considered.     

Asymmetric ACD models: Introducing price information in ACD models

This paper proposes an asymmetric autoregressive conditional duration (ACD) model, which extends the ACD model of Engle and Russell (1998). The asymmetry consists of letting the duration process depend on the state of the price process. If the price has increased, the parameters of the ACD model can differ from what they are if the price has decreased. The model is applied to the bid-ask quotes of two stocks traded on the NYSE and the evidence in favour of asymmetry is strong. Information effects (Easley and O'Hara 1992) are also empirically relevant. As the model is a transition model for the price process, it delivers `market forecasts' of where prices are heading. A trading strategy based on the model is implemented using tick-by-tick data.While remaining responsible for any error in this paper, the authors would like to thank R. Anderson, G. Le Fol, C. Gouriéroux, J. Jasiak, W. Pohlmeier, A. Roell, O. Scaillet, S. Wei and three anonymous referees for useful remarks and suggestions on previous versions. The authors would also like to thank A. Ruttiens from KBC-CBC for useful discussions on practical issues related to trading. Support of the European Commission Human Capital and Mobility Program through the network `Econometric inference using simulation methods' is gratefully acknowledged. This paper presents research results of the Belgian Program on Interuniversity Poles of Attraction initiated by the Belgian State, Prime Minister's Office, Science Policy Programming. The scientific responsibility is assumed by the authors.     

Reinterpreting the performance of immigrant wages from panel data

Immigrants differ from the native born in terms of unobserved factors, such as motivation, and observed factors, including those related to the interruption of labour market activity and earning capacity, which may bias estimates of immigrant integration. Using panel data from the Survey of Labour and Income Dynamics, we show that using potential experience, rather than actual experience, exaggerates estimates of the disruption and recovery caused by immigration. More importantly, we find support for omitted variables bias, arising from unobserved fixed effects. Instrumental variable estimates for both pooled and separate samples of immigrant and native born men demonstrate a wage disadvantage for immigrants upon entry that persists through their lifetime. Standard estimates of a modest wage advantage for the children of immigrants also suffer from omitted variables bias arising from unobservables. Contrary to most of the literature to date, our instrumental variable estimates which allow for unobservable fixed effects suggest that immigrants never catch up to otherwise comparable native born workers, but their children do just as well. We would like to thank Statistics Canada for permitting access to the data and solving associated technical problems, the Prairier Centre for Research on Immigration and Integration for financial assistance, and Peter Schnabl for excellent research assistance. An earlier version of this paper was presented to the 10th International Conference on Panel Data in Berlin, July 5–6, 2002. The authors take sole responsibility for errors, omissions and interpretation of the data.     

Interbank Relationships and the Credit Channel in Germany

A crucial condition for the existence of a credit channel through bank loansis that monetary policy should be able to change bank loan supply. This papercontributes to the discussion on this issue by presenting empirical evidence fromdynamic panel estimations based on a dataset that comprises individual balancesheet information on all German banks. It shows that the average bank reduces itslending more sharply in reaction to a restrictive monetary policy measure the lowerits ratio of short-term interbank deposits to total assets. A dependence on its size canonly be found if explicitly controlled for this dominating effect. Overall, the evidenceis compatible with the existence of a credit channel but the results indicate that it is weakened by the network structures that exist in the German banking system.     

数据拟合回归方法的探讨

建立回归方程进行分析预测,大多采用人们熟知的"最小二乘法"。但是"最小二乘法"有本质上的一个弱点,由于系数矩阵的病态性质:系数矩阵中的最大值相对很大,系数矩阵的行列式值相对较小,在求解方程时,系数中的微小变化可能导致结果不稳定,预测效果不佳甚至失真。本文另辟蹊径,与"最小二乘法"的拟合原则不同,我们是最小化误差绝对值中的最大值。该法克服了"最小二乘法"有时不稳定的缺陷,继而提高拟合精度,使预测效果更精确。     

房地产价格波动与投机行为--对中国14城市的实证研究

本文首先构建了适合中国的房地产投机理论模型,然后利用该模型对中国14个城市房地产价格波动与投机行为的关系进行实证研究.对14城市的时间序列数据研究结果表明各城市房地产投机水平都很高,个别城市更加突出.通过对14城市的截面数据分析,可以看出可支配收入对房地产价格没有显著影响,这些城市房地产价格的上升,主要是由投机来推动的,而且整体投机度非常高,说明房地产价格极大的偏离长期均衡值,市场出现了非理性繁荣,政府和产业部门应该采取适当的措施,积极引导消费者和投资者,为建立均衡的房地产市场提供良好的宏观和微观环境.     

Is Private R & D Spending Sensitive to Its Price? Empirical Evidence on Panel Data for Italy

In this paper empirical evidence is presented on theelasticity of private R & D spending on its price. Acensored panel-data regression model with random effectsis applied to a balanced panel of 726 Italian firms overthe 1992–1997 period. Implied estimates point out thatItalian firms' response to policy measures (including taxcredits), aimed at reducing the user cost of R & D capital,is likely to be substantial (1.50–1.77). Furthermore, wealso find that the elasticity of R & D spending is higherin recession (2.01) than in expansion (0.87).     

城市森林固定CO2价值评估

本文利用我国第四次森林资源清样调查资料对上海市森林植被碳储量和碳密度的估算结果,结合植物的光合作用过程计算上海市森林固定CO2的能力,采用造林成本法对上海市森林固定CO2的价值进行了评估。笔者希望通过该研究激发人们保护城市森林以维持城市生态系统可持续发展的积极性,并为合理地进行城市森林管理和城市绿化提供参考。另外,利用系统连续的森林资源清样调查资料评估森林植被的固碳价值,能够对我国更好地履行《联合国气候变化框架公约(》UNFCCC),并为依据《京都议定书(》KyotoProtocol)进行环境外交提供帮助。     

自动化仓库系统可靠性模糊评价

通过对影响自动化仓库系统各种因素的分析，用模糊综合评价方法对自动化仓库系统的可靠性进行评价。     

Comparison of Hypothesis Testing and Bayesian Model Selection

The main goal of both Bayesian model selection and classical hypotheses testing is to make inferences with respect to the state of affairs in a population of interest. The main differences between both approaches are the explicit use of prior information by Bayesians, and the explicit use of null distributions by the classicists. Formalization of prior information in prior distributions is often difficult. In this paper two practical approaches (encompassing priors and training data) to specify prior distributions will be presented. The computation of null distributions is relatively easy. However, as will be illustrated, a straightforward interpretation of the resulting p-values is not always easy. Bayesian model selection can be used to compute posterior probabilities for each of a number of competing models. This provides an alternative for the currently prevalent testing of hypotheses using p-values. Both approaches will be compared and illustrated using case studies. Each case study fits in the framework of the normal linear model, that is, analysis of variance and multiple regression.