Publication: Some applications of the useful mutual information
Full text at PDC
Advisors (or tutors)
The useful mutual information statistic is considered and its asymptotic properties are investigated either considering random sampling or stratified random sampling. Under different assumptions, it is shown that the asymptotic distribution of this statistic is either normal or a linear form in chi(2) variables. Finally, its application to testing hypotheses of independence in multinomial populations is given.
G. A. Miller, Note on the bias of information estimates, in Information Theory in Psycology ( H. Quastler, Ed.), 1955, pp. 95-100. G. Longo, A Quantitative Qualitative Measure of Information, Springer Verlag, 1972. B.D. Sharma, M. Mohan, and J. Mitter, Informational Approach to Aggregation Problem in Financial Statements, personal communication 1977. C.F. Picard, Weighted probabilistic information measures, Comb. Inform. Syst. Sci. 4:343-356 (1979). K. Zografos, Asymptotic properties of phi-divergence statistic and its application in contingence tables, Internat. J. Math. Statist. Sci., to appear. L. Pardo, M. Salicrú, M. L. Menéndez, and D. Morales, The phi-divergence statistic in bivariate multinomial populations including stratification, Metrika, to appear. K.V. Marclia, J. T. Kent, and J. M. Bibby, Multivariate Analysis, Academic Press, 1982. S. Kotz, N. M. Johnson, and D. W. Boid, Series representation of quadratic forms in normal variables. I. Central Case, AMS 823-837 (1967). J.N.K. Rao and A. J. Scott, The analysis of categorical data from complex sample surveys Chi-squaxed tests for goodness of fit and independence in two way tables, J. Araer. Star. Assoc. 76:221-230 (1981). C.R. Rao, Linear Statistical Inference and its Applications, Wiley, New York, 1973.