Statistical Inference Vol. 70

By George Casella, R. Berger. 1990

How researchers see this paper

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

Interpreting support vector machine models for multivariate group wise analysis in neuroimaging (2015, positive): “While the p-values produced by our inference framework may be applied to improve the interpretability of neuroimaging based SVM classifier it should not be confused for multivariate statistical testing frameworks like MANOVA Casella and Berger (2002) Srivastava and Du (2008) Fujikoshi et al. (2004) ”
Can you Trust the Trend?: Discovering Simpson's Paradoxes in Social Data (2018, methodology): “…outcomes we use the logistic regression linear model of the form parameters α and β are fitted using Maximum likelihood, while test of the null hypothesis H 0 : β = 0 is performed using the Likelihood Ratio Test [6]. eleven variables in Stack Exchange data, result in 110 possible Simpson’s pairs.”
Demographics of News Sharing in the U.S. Twittersphere (2017, methodology): “To control the effect we use a Chi-square test [9] to find which topics are written significantly more by Female (or Male) journalists and which topics are significantly more shared by Female (or Male) spreaders.”
Cardinality Estimators do not Preserve Privacy (2018, background): “Theorem 12 shows, for all of them, that there are some users whose privacy loss is signiﬁcant.”