Returns the Fligner-Kileen test for homogeneous scales for two-samples. Also estimates of ratio of scales based on the logs of folded median-aligned samples and a corresponding confidence interval is computed. fk.test computes the value of the statistic based on squared-normal scores following the optimal (for normal errors) such test described in Section 2.10 of Hettmansperger and McKean (2011). Hence, it will differ from the core R routine fligner.test; see the discussion in Section 3.3 of Kloke and McKean (2014)/Section 3.5 of Kloke and McKean (2024).
fk.test(x,y,alternative = c("two.sided", "less", "greater"),conf.level = 0.95)
chi-squared test statistic
p-value of the test
vector of estimates of ratio of scales
table of confidence intervals
vector of first sample responses
vector of second sample responses
alternative indicator for hypotheses
confidence coefficient for the returned confidence intervals
John Kloke, Joseph McKean
Returns the Fligner-Kileen test for the two-sample scale problem.
Kloke, J. and McKean, J.W. (2014), Nonparametric statistcal methods using R, Boca Raton, FL: Chapman-Hall.
Kloke, J. and McKean, J.W. (2024), Nonparametric statistcal methods using R, Second Edition, Boca Raton, FL: Chapman-Hall.
Hettmansperger, T.P. and McKean J.W. (2011), Robust Nonparametric Statistical Methods, 2nd ed., New York: Chapman-Hall.
fkk.test
x<-rnorm(18)
y<-rnorm(22)*3
fk.test(x,y)
Run the code above in your browser using DataLab