摘要：This paper introduces a new way of investigating linear and nonlinear Granger causality between exports, imports and economic growth in France over the period 1961_2006 with using geostatistical models (kiriging and Inverse distance weighting). Geostatistical methods are the ordinary methods for forecasting the locatins and making map in water engineerig, environment, environmental pollution, mining, ecology, geology and geography. Although, this is the first time which geostatistics knowledge is used for economic analyzes. In classical econometrics there do not exist any estimator which have the capability to find the best functional form in the estimation. Geostatistical models investigate simultaneous linear and various nonlinear types of causality test, which cause to decrease the effects of choosing functional form in autoregressive model. This approach imitates the Granger definition and structure but improve it to have better ability to investigate nonlinear causality. Taking into account the results of linear and non linear (using geostatistical method) causality analysis, results give strong evidence that there was causality running from GDP to trade. Additionally, the nonlinear causality analysis also leads to the conclusion that export was a causal factor for import. Our result supports the GLE model in France.
关键词：Granger causality; Exports; Imports; Economic growth; Geostatistical ; model
