Bayesianly Justifiable and Relevant Frequency Calculations for the Applied Statistician

By Donald B. Rubin. 1984

Subfields: Probabilistic & Bayesian Methods

How researchers see this paper

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

In Probabilistic & Bayesian Methods

Citations in the corpus

The frontier of simulation-based inference (2019, methodology): “Alternatively, a technique known as approximate Bayesian computation (ABC) (2, 3) compares the observed and simulated data based on some distance measure involving the summary statistics.”
Time Series Deconfounder: Estimating Treatment Effects over Time in the Presence of Hidden Confounders (2019, methodology): “This condition can be assessed by extending predictive model checking (Rubin, 1984) to the time-series setting.”
Edward: A library for probabilistic modeling, inference, and criticism (2016, methodology): “For criticism, Edward provides methods from scoring rules (Winkler, 1996) and predictive checks (Box, 1980; Rubin, 1984).”
Probabilistic forecasts, calibration and sharpness (2007, background): “Rubin (1984), pages 1161 and 1160, argued that ‘the probabilities attached to Bayesian statements do have frequency interpretations that tie the statements to verifiable real world events’.”
Probabilistic topic models (2012, unknown): “Posterior predictive checks • Rubin (1984) located the data y in the posterior p(y |y,M).”
Understanding predictive information criteria for Bayesian models (2013, methodology): “The ﬁt of model to data can be assessed using posterior predictive checks (Rubin, 1984), prior predictive checks (when evaluating potential replications involving new parameter values), or, more generally, mixed checks for hierarchical models (Gelman, Meng, and Stern, 2006).”
Penalising Model Component Complexity: A Principled, Practical Approach to Constructing Priors (2014, methodology): “An interesting, but computationally intensive, method for reasserting the role of an “expert” into a class of ad hoc priors is the method of calibrated Bayes (Rubin, 1984; Browne and Draper, 2006), where the hyper-parameters in the prior are chosen to ensure that, under correct model specification, ”
Bayesian graphical modelling: a case‐study in monitoring health outcomes (2002, methodology): “Details of the data are provided by Gibson et al. (1997)Ðhere we group the cases by age of diagnosis (20±34, 35±49, 50±69 and 70 years or older), year (1971±1995) and place of residence (coded as one of the eight district health authorities existing in East Anglia until 1992).”
Bayes factors (1995, background): “Many Bayesians do, however, see a useful role for goodness-of--t tests (Dempster, 1971; Box, 1980; Rubin, 1984).”
A Bayesian test for periodic signals in red noise (2009, background): “This tail area probability does not depend on the unknown value of parametersθ, and is often called theposterior predictive p-value (see Rubin 1984; Meng 1994; Gelman et al. 1996, 2004; Protassov et al. 2002).”
The Prior Can Often Only Be Understood in the Context of the Likelihood (2017, background): “Regularization, even if applied in a Bayesian context, is a frequentist goal [4] in that its success is quantified in terms of the statistical properties of the inferences from an ensemble of possible measurements.”
Suggestions for presenting the results of data analyses (2001, background): “One particularly simple and easily implemented check is a posterior predictive check (Rubin 1984).”