Generalized Correlation and Kernel Causality with Applications in Development Economics
Hrishikesh D. Vinod
Fordham University - Department of Economics
November 24, 2013
New generalized correlation measures, GMC(Y|X), of 2012 use Kernel regressions to overcome the linearity of Pearson correlation coefficients. A new matrix of generalized correlation coefficients is such that when |r*(i,j)|>|r*(j,i)|, it is more likely that the column variable Xj is what Granger called the "instantaneous cause" or what we call "kernel cause" of the row variable Xi. New partial correlations ameliorate confounding. Various examples and simulations support robustness of new causality. We include bootstrap inference, robustness checks based on dependence between regressor and error and out-of-sample forecasts. Data for 198 countries on nine development variables support growth policy over redistribution and Deaton's criticism of foreign aid. Potential applications include Big Data, since my R code is available in the supplementary material.
Number of Pages in PDF File: 36
Keywords: bootstrap, poverty reduction, foreign aid, entrepreneurship
JEL Classification: O57, O10, C15, C31working papers series
Date posted: November 7, 2013 ; Last revised: January 4, 2015
© 2015 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo1 in 0.329 seconds