Uncertain significance test for regression coefficients with application to regional economic analysis
Tingqing Ye, Baoding Liu
Abstract
This paper presents a statistical tool of uncertain significance test that uses uncertainty theory to test whether certain prespecified regression coefficients can be regarded as zero. A numerical example is given to illustrate how to test the significance of regression coefficients in an uncertain regression model. In order to compare uncertain significance test with stochastic significance test, both of these significance testing approaches are applied in studying the relationship between GDP and four indicators, including urban population scale, volume of foreign trade, fiscal expenditure, and water resource. The results show that uncertain significance test is more appropriate than stochastic significance test.
Topics & Concepts
Test (biology)Statistical significanceEconometricsRegression analysisRegressionStatisticsLinear regressionPopulationStatistical hypothesis testingMathematicsMedicineGeologyEnvironmental healthPaleontologyFuzzy Systems and OptimizationMulti-Criteria Decision Making