翻訳と辞書 Words near each other ・ Maxima of Rome ・ Maxima, Kumasi ・ Maximag ・ Maximal ・ Maximal (Transformers) ・ Maximal arc ・ Maximal common divisor ・ Maximal compact subgroup ・ Maximal Crazy ・ Maximal element ・ Maximal ergodic theorem ・ Maximal evenness ・ Maximal function ・ Maximal ideal ・ Maximal independent set ・ Maximal information coefficient ・ Maximal lotteries ・ Maximal munch ・ Maximal pair ・ Maximal semilattice quotient ・ Maximal set ・ Maximal subgroup ・ Maximal torus ・ Maximal-ratio combining ・ Maximalism ・ Maximalist! (band) ・ Maximally informative dimensions ・ Maximally stable extremal regions ・ Maximaphily ・ Maximato Dictionary Lists mini英和辞書 mini和英辞書 Webster 1913 Latin-English FOLDOC Wikipedia English ウィキペディア

翻訳と辞書　辞書検索 [ 開発暫定版 ] スポンサード リンク Maximal information coefficient ： ウィキペディア英語版 Maximal information coefficient In statistics, the maximal information coefficient (MIC) is a measure of the strength of the linear or non-linear association between two variables ''X'' and ''Y''. The MIC belongs to the maximal information-based nonparametric exploration (MINE) class of statistics. In a simulation study, MIC outperformed some selected low power tests,〔Reshef et al. 2011〕 however concerns have been raised regarding reduced statistical power in detecting some associations in settings with low sample size when compared to powerful methods such as distance correlation and HHG.〔(A consistent multivariate test of association based on ranks of distances )〕 Comparisons with these methods, in which MIC was outperformed, were made in 〔Noah Simon and Robert Tibshirani, (Comment on “Detecting Novel Associations in Large Data Sets” by Reshef et al., Science Dec. 16, 2011 )〕 and.〔(Comment on "Detecting Novel Associations in Large Data Sets" )〕 It is claimed〔 that MIC approximately satisfies a property called ''equitability'' which is illustrated by selected simulation studies.〔 It was later proved that no non-trivial coefficient can exactly satisfy the ''equitability'' property as defined by Reshef et al.,〔〔(Equitability, mutual information, and the maximal information coefficient by Justin B. Kinney, Gurinder S. Atwal, arXiv Jan. 31, 2013 )〕 although this result has been challenged.〔(R2-equitability is satisfiable by Ben Murrell, Daniel Murrell, and Hugh Murrell, Proceedings of the National Academy of Sciences, May. 27, 2014 )〕 Some criticisms of MIC are addressed by Reshef et al. in further studies published on arXiv.〔(Equitability Analysis of the Maximal Information Coefficient, with Comparisons by David Reshef, Yakir Reshef, Michael Mitzenmacher, Pardis Sabeti, arXiv Jan. 27, 2013 )〕 == Overview == The maximal information coefficient uses binning as a means to apply mutual information on continuous random variables. Binning has been used for some time as a way of applying mutual information to continuous distributions; what MIC contributes in addition is a methodology for selecting the number of bins and picking a maximum over many possible grids. The rationale is that the bins for both variables should be chosen in such a way that the mutual information between the variables be maximal. That is achieved whenever $\backslash mathrm\backslash left(X\_b\backslash right)=\backslash mathrm\backslash left(Y\_b\backslash right)=\backslash mathrm\backslash left(X\_b,Y\_b\backslash right)$.〔The "b" subscripts have been used to emphasize that the mutual information is calculated using the bins〕 Thus, when the mutual information is maximal over a binning of the data, we should expect that the following two properties hold, as much as made possible by the own nature of the data. First, the bins would have roughly the same size, because the entropies $\backslash mathrm(X\_b)$ and $\backslash mathrm(Y\_b)$ are maximized by equal-sized binning. And second, each bin of ''X'' will roughly correspond to a bin in ''Y''. Because the variables X and Y are reals, it is almost always possible to create exactly one bin for each (''x'',''y'') datapoint, and that would yield a very high value of the MI. To avoid forming this kind of trivial partitioning, the authors of the paper propose taking a number of bins $n\_x$ for ''X'' and $n\_y$ whose product is relatively small compared with the size N of the data sample. Concretely, they propose: $n\_x\backslash times\; n\_y\; \backslash leq\; \backslash mathrm^$ In some cases it is possible to achieve a good correspondence between $X\_b$ and $Y\_b$ with numbers as low as $n\_x=2$ and $n\_y=2$, while in other cases the number of bins required may be higher. The maximum for $\backslash mathrm(X\_b;Y\_b)$ is determined by H(X), which is in turn determined by the number of bins in each axis, therefore, the mutual information value will be dependent on the number of bins selected for each variable. In order to compare mutual information values obtained with partitions of different sizes, the mutual information value is normalized by dividing by the maximum achieveable value for the given partition size. Entropy is maximized by uniform probability distributions, or in this case, bins with the same number of elements. Also, joint entropy is minimized by having a one-to-one correspondence between bins. If we substitute such values in the formula $I(X;Y)=H(X)+H(Y)-H(X,Y)$, we can see that the maximum value achieveable by the MI for a given pair $n\_x,n\_y$ of bin counts is $\backslash log\backslash min\backslash left(n\_x,n\_y\backslash right)$. Thus, this value is used as a normalizing divisor for each pair of bin counts. Last, the normalized maximal mutual information value for different combinations of $n\_x$ and $n\_y$ is tabulated, and the maximum value in the table selected as the value of the statistic. It is important to note that trying all possible binning schemes that satisfy $n\_x\backslash times\; n\_y\; \backslash leq\; \backslash mathrm^$ is computationally unfeasible even for small n. Therefore in practice the authors apply a heuristic which may or may not find the true maximum. 抄文引用元・出典: フリー百科事典『 ウィキペディア（Wikipedia）』 ■ウィキペディアで「Maximal information coefficient」の詳細全文を読む スポンサード リンク 翻訳と辞書 : 翻訳のためのインターネットリソース Copyright(C) kotoba.ne.jp 1997-2016. All Rights Reserved.