Use Hoeffding's D to Quantify and Test Non-monotonic Dependence
Unlike SpearmanRho, KendallTau, and Pearsons's Correlation, HoeffdingD can be used to detect a wide variety of dependence structures beyond monotonic association.
| In[1]:= |  X |
| In[2]:= |  X |
| Out[2]= |  |
Test whether Hoeffding's D is significantly different from zero.
| In[3]:= |  X |
| Out[3]= |  |
Here is a collection of datasets with different dependency structures.
| Out[30]= |  |
The Hoeffding D statistics and test conclusions for the data grid follow.
| In[4]:= |  X |
| Out[4]//TableForm= |
 |