Robust tests for equality of variances

By H. Levene. 1961

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Invariant Causal Prediction for Nonlinear Models (2017, methodology): “In step 4, we use both approaches described above, using (a) ks.test() from the stats package in R (Conover, 1971) and (b) wilcox.test() and levene.test() (the latter being contained in the lawstat package in R (Levene, 1960; Gastwirth et al., 2015)).”
On the Stability of Fine-tuning BERT: Misconceptions, Explanations, and Strong Baselines (2020, methodology): “To further validate our claim about the fine-tuning stability, we run Levene’s test (Levene, 1960) to check the equality of variances for the distributions of the final performances on each dataset.”
Causal Transfer in Machine Learning (2015, methodology): “For Gaussian data, one can use a Levene test [Levene, 1960] to test whether the residuals have the same variance in all tasks; their means are zero as long as an intercept is included in the regression model.”
Evaluation metrics and statistical tests for machine learning (2024, methodology): “55 If the data is not normally distributed, the possible alternatives for the F-test include Barlett’s test ( bartlett in scipy.stats in Python and bartlett.test in stats in R) and Levene’s test 56 ( levene in scipy.stats in Python and leveneTest in the package car 57 in R).”
Uncertainty analysis of statistical downscaling methods (2006, unknown): “The computational method employed here for Levene’s Test is a modification of Levene’s original procedure by Brown and Forsythe (1974).”
Distance‐Based Tests for Homogeneity of Multivariate Dispersions (2006, methodology): “Brown and Forsythe (1974) suggested that medians be used instead of means for Levene’s test, and this approach was found to be the most robust and useful of 56 possible tests for homogeneity in extensive simulations done by Conover, Johnson, and Johnson (1981).”
The two-sample t test: pre-testing its assumptions does not pay off (2011, methodology): “As a pre-test of normality, we use the Kolmogorov–Smirnov-test (Kolmogorov 1933; Smirnov 1939) and as a pre-test of variance homogeneity the Levene-test ( Levene 1960; this because according to Rasch and Guiard 2004 ,t heF test is very sensitive against non-normality and has already been replaced i”