WHICH PART OF THE SAMPLE CONTAINS THE INFORMATION?

By J. Tukey. 1965

How researchers see this paper

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

Autoregressive Quantile Networks for Generative Modeling (2018, background): “Indeed, this quantity, known as the sparsity function (Tukey, 1965) or the quantile-density function (Parzen, 1979) plays a central role in the analysis of quantile regression models (Koenker, 1994).”
Higher criticism for detecting sparse heterogeneous mixtures (2004, background): “Which part of the sample contains the information? Underlying our results is a set of insights about “where to look” for evidence against H0 [30], how the evidence may not be in the “obvious” place and how the adaptation in HC∗ automatically ensures that the best evidence will be included in making ”
Functional data analysis for density functions by transformation to a Hilbert space (2016, background): “This concept goes back to [37] and [46].”
The Influence Curve and Its Role in Robust Estimation (1974, unknown): “also [35]).”
Forecasting of density functions with an application to cross-sectional and intraday returns (2019, background): “Of these three, Qt is the least constrained, and it is easy to see that the so-called quantile density (Jones, 1992; Parzen, 1979; Tukey, 1965) qt = Q ′t is only constrained to be nonnegative.”
ESTIMATING DENSITIES , QUANTILES , QUANTILE DENSITIES AND DENSITY QUANTILES (2004, methodology): “We follow Parzen's (1979) terminology in calling q the quantile density function; Tukey (1965) called it the sparsity function.”