July 2016
Beginner to intermediate
462 pages
9h 14m
English
Pearson's r, named after its developer Karl Pearson (1896), measures linear correlation between two variables. Let's look at the following equations:

(3.13) defines the coefficient and (3.14) describes the
Fisher transformation used to compute confidence intervals. (3.15) gives the standard error of the correlation. (3.16) is about the z-score of the Fisher transformed correlation. If we assume a normal distribution, we can use the z-score to compute confidence intervals. Alternatively, we can bootstrap by resampling pairs of values with replacement. Also, the scipy.stats.pearsonr() function returns ...
Read now
Unlock full access