Tests of statistical hypotheses concerning several parameters when the number of observations is large

By A. Wald. 1943

How researchers see this paper

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

In Statistics and Probability

Citations in the corpus

Status Conflict in Groups (2012, methodology): “To determine whether the addition of variables in the hierarchical analysis improves the model fit in robust regression analyses, we conduct Wald tests, which calculate whether or not the coefficients of the added variables at each step are simultaneously zero (Wald 1943).”
Inference Functions and Quadratic Score Tests (2003, methodology): “It can be shown that Rao’s score test is asymptotically equivalent to the likelihood ratio test (Neyman and Pearson, 1933) and the Wald (1943) test under both the null and the Pitman alternative hypotheses (Serfling, 1980, page 156), and all follow the chi-squared distribution asymptotically.”
Minerva Unbound: Knowledge Stocks, Knowledge Flows and New Knowledge Production (2007, unknown): “We note that all the coefficients but one in Table 3 are highly significant and do make comparative evaluations of goodness of fit using Wald (1943) log-likelihood tests.”
A MATLAB toolbox for Granger causal connectivity analysis. (2010, unknown): “2 It is also known that the maximum likelihood estimator F̂ will have (asymptotically for large samples) a 2-distribution under the null hypothesis F = 0 (Granger, 1963; Whittle, 1953) and a non-central 2-distribution under the alternative hypothesis F > 0 (Geweke, 1982; Wald, 1943).”
Categorical Data Analysis: Away from ANOVAs (transformation or not) and towards Logit Mixed Models (2008, methodology): “The standard errors are used to calculate Wald’s z-score (henceforth Wald’s Z, Wald, 1943) in the fourth column by dividing the coefficient estimate by the estimate for its standard error.”
Are Robust Standard Errors the Best Approach for Interval Estimation With Nonnormal Data in Structural Equation Modeling? (2018, methodology): “In a more general form, including a possible function of model parameters, this same test can be computed via a Wald-based statistic (Wald, 1943), 3 Except for distributional violations, all models in this article are correctly specified, and so use of expected information is justified.”
Robust Inference for Generalized Linear Models (2001, methodology): “These conditions are obtained by replacing logf 4z1 ‚5 by QM 4z1‚5 in the corresponding classical results for the likelihood ratio test (Rao 1973; Wald 1943).”
Applying a Score Confidence Interval to Aiken's Item Content-Relevance Index (2004, background): “That is, because V is not normally distributed, traditional confidence intervals for a population mean, such as the Wald interval (Wald, 1943) given by most introductory statistics texts as will lead to inaccurate results.”
Marginal models for categorical data (2002, methodology): “The Wald statistic (Wald, 1943) can be used for testing goodness-offit even without estimated frequencies.”