In this paper we examine 5 indexes (the two Yule’s indexes, the chi square, the odds ratio and an elementary index) of a two-by-two table, which estimate the correlation coefficient ρ in a bivariate Bernoulli distri...In this paper we examine 5 indexes (the two Yule’s indexes, the chi square, the odds ratio and an elementary index) of a two-by-two table, which estimate the correlation coefficient ρ in a bivariate Bernoulli distribution. We will find the compact expression of the influence functions, which allow the quantification of the effect of an infinitesimal contamination of the probability of any pair of attributes of the bivariate random variable distributed according to the above-mentioned model. We prove that the only unbiased index is the chi square. In order to determine the indexes, which are less sensitive to contamination, we obtain the expressions of three synthetic measures of the influence function, which are the maximum contamination (gross sensitivity error), the mean square deviation and the variance. These results, even if don’t allow a definitive assessment of the overall optimum properties of the five indexes, as not all of them are unbiased, nevertheless they allow to appreciating the synthetic entity of the effect of the contaminations in the estimation of the parameter ρ of the bivariate Bernoulli distribution.展开更多
This study uses <span style="font-family:Verdana;">an empirical</span><span style="font-family:Verdana;"> analysis to quantify the downstream analysis effects of data pre-processi...This study uses <span style="font-family:Verdana;">an empirical</span><span style="font-family:Verdana;"> analysis to quantify the downstream analysis effects of data pre-processing choices. Bootstrap data simulation is used to measure the bias-variance decomposition of an empirical risk function, mean square error (MSE). Results of the risk function decomposition are used to measure the effects of model development choices on </span><span style="font-family:Verdana;">model</span><span style="font-family:Verdana;"> bias, variance, and irreducible error. Measurements of bias and variance are then applied as diagnostic procedures for model pre-processing and development. Best performing model-normalization-data structure combinations were found to illustrate the downstream analysis effects of these model development choices. </span><span style="font-family:Verdana;">In addition</span><span style="font-family:Verdana;">s</span><span style="font-family:Verdana;">, results found from simulations were verified and expanded to include additional data characteristics (imbalanced, sparse) by testing on benchmark datasets available from the UCI Machine Learning Library. Normalization results on benchmark data were consistent with those found using simulations, while also illustrating that more complex and/or non-linear models provide better performance on datasets with additional complexities. Finally, applying the findings from simulation experiments to previously tested applications led to equivalent or improved results with less model development overhead and processing time.</span>展开更多
文摘In this paper we examine 5 indexes (the two Yule’s indexes, the chi square, the odds ratio and an elementary index) of a two-by-two table, which estimate the correlation coefficient ρ in a bivariate Bernoulli distribution. We will find the compact expression of the influence functions, which allow the quantification of the effect of an infinitesimal contamination of the probability of any pair of attributes of the bivariate random variable distributed according to the above-mentioned model. We prove that the only unbiased index is the chi square. In order to determine the indexes, which are less sensitive to contamination, we obtain the expressions of three synthetic measures of the influence function, which are the maximum contamination (gross sensitivity error), the mean square deviation and the variance. These results, even if don’t allow a definitive assessment of the overall optimum properties of the five indexes, as not all of them are unbiased, nevertheless they allow to appreciating the synthetic entity of the effect of the contaminations in the estimation of the parameter ρ of the bivariate Bernoulli distribution.
文摘This study uses <span style="font-family:Verdana;">an empirical</span><span style="font-family:Verdana;"> analysis to quantify the downstream analysis effects of data pre-processing choices. Bootstrap data simulation is used to measure the bias-variance decomposition of an empirical risk function, mean square error (MSE). Results of the risk function decomposition are used to measure the effects of model development choices on </span><span style="font-family:Verdana;">model</span><span style="font-family:Verdana;"> bias, variance, and irreducible error. Measurements of bias and variance are then applied as diagnostic procedures for model pre-processing and development. Best performing model-normalization-data structure combinations were found to illustrate the downstream analysis effects of these model development choices. </span><span style="font-family:Verdana;">In addition</span><span style="font-family:Verdana;">s</span><span style="font-family:Verdana;">, results found from simulations were verified and expanded to include additional data characteristics (imbalanced, sparse) by testing on benchmark datasets available from the UCI Machine Learning Library. Normalization results on benchmark data were consistent with those found using simulations, while also illustrating that more complex and/or non-linear models provide better performance on datasets with additional complexities. Finally, applying the findings from simulation experiments to previously tested applications led to equivalent or improved results with less model development overhead and processing time.</span>