Repeat Examples 14.4 and 14.5 on some other available data set in any R package or other source.

Transcribed Image Text:

Example 14.4. This example compares dCor and Pearson correlation in exploratory data analysis. Consider the Freedman [1975] census data from [United States Bureau of the Census, 1970] on crime rates in U. S. metropoli- tan areas with 1968 populations of 250,000 or more. The data set is available from the car package for R. [Fox and Weisberg, 2019], and contains four nu- meric variables: population (total 1968, in thousands) nonwhite (percent nonwhite population, 1960) density (population per square mile, 1968) crime (crime rate per 100,000, 1969) The 110 observations contain missing values. The data analyzed are the 100 cities with complete data. Pearson and dCor statistics Rn are shown in Table 14.1. Note that there is a significant association between crime and population density measured by dCor, which is not significant when measured by p. TABLE 14.1 Pearson correlation and distance correlation statistics for the Freedman data of Example 14.4. Significance at 0.05, 0.01, 0.001 for the corresponding tests is indicated by *, **, ***, respectively. nonwhite population 0.070 nonwhite density Pearson dCor density crime nonwhite density crime 0.368*** 0.396 ** 0.260* 0.615*** 0.422* 0.002 0.294** 0.194 0.112 Analysis of this data continues in Example 14.5. 0.385*** 0.250*