---------- ---------- ---------- ---------- ---------- ---------- ---------- ----------
# [  source("http://macosa.dima.unige.it/r.R")  ]
# Comment on "cor.test".
#
H = c(15,20,39,52); F = c(260,380,710,990)
cor.test(H,F, conf.level=0.90)
# 
# Pearson's product-moment correlation
# 
# data:  H and F
# t = 28.471, df = 2, p-value = 0.001231
# alternative hypothesis: true correlation is not equal to 0
# 90 percent confidence interval:
#  0.9674734  0.9999541
# sample estimates:
#       cor 
# 0.9987686 
# 
# Pearson: test based on Pearson's method
# df: "degrees of freedom" (amount of data decreased by two)
# p-value; probability of obtaining this "cor" with such a quantity of data in the
# hypothesis that the two variables were totally unrelated (ie cor was actually 0)
# (1 - 0.0012314 = 0.9987686)
# t is cor(H,F)*sqrt(n-2)/sqrt(1-cor(H,F)^2) where n = length(H); it is used in
# estimating some parameters