Parametric multiple groups

ANOVA:

  • Null hypothesis (H0): The means of the results of two or more algorithms are the same.
  • Data should satisfy: normality, homocedasticity and independence. [?]

Bonferroni:

  • Null hypothesis (H0): The mean of the results of the "Algorithm A" and "Algorithm B" is equal (compares all algorithms each other).
  • Is applied if: ANOVA indicates that there are differences.
References

  • ANOVA: Box, G. E. (1953). Non-normality and tests on variance. Biometrika, 40, 318–335.

  • Bonferroni-Dunn: Dunn, O. J. (1961). Multiple comparisons among means. Journal of the American Statistical Association, 56, 52–64.