The direct use of likelihood for significance testing
The DOI matches its Crossref record; 5 of the 5 references it cites were found in a registry or an index.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
On the mathematical foundations of theoretical statistics
Fisher, R. A. (1922) · Philosophical Transactions of the Royal Society of London. Series A, Containing Papers of a Mathematical or Physical Character · doi:10.1098/rsta.1922.0009 · Cited by 4,039 · cited 1×
“(Fisher, 1922) The connection between probability and suspicion is that the occurrence of an event of small probability raises doubts, the familiar example being the ordinary signi®cance test whose interpretation was phrased by Fisher as follows: The force with which such a conclusion is supported…”
BackgroundStatistical Methods and Scientific Inference.
Bartlett, M. S., & Fisher, R. A. (1957) · Biometrika · doi:10.2307/2333276 · Cited by 1,278
Statistical Methods and Scientific Inference.
Bailey, N. T. J., & Fisher, R. (1957) · Journal of the Royal Statistical Society. Series A (General) · doi:10.2307/2342557 · Cited by 760
Philosophy, science, and method : essays in honor of Ernest Nagel
Nagel, E., Morgenbesser, S., Suppes, P., & White, M. (1969) · St Martin's Press eBooks · Cited by 443
On the Difficulties Inherent in Fisher's Fiducial Argument
Dempster, A. P. (1964) · Journal of the American Statistical Association · doi:10.1080/01621459.1964.10480700 · Cited by 39 · cited 1×
“In an earlier attempt to focus attention on the contrast between the two forms of direct inference (Dempster, 1964), I introduced the terms postdictive and predictive, feeling then as now that a distinctive new term was needed to dramatize the retrospective mode.”
BackgroundSelf-citation
Research works printed but not linked: 1
- Edwards, A. W. F. (1972) Likelihood. Cambridge University Press, Cambridge.
Cited by
Understanding predictive information criteria for Bayesian models
Gelman, A., Hwang, J., & Vehtari, A. (2014) · Statistics and Computing · doi:10.1007/s11222-013-9416-2 · Cited by 2,104
“Figure 2 is reminiscent of the direct likelihood methods of Dempster (1974) and Aitkin (2010).”
MethodDias, S., Sutton, A. J., Ades, A. E., & Welton, N. J. (2013) · Medical Decision Making · doi:10.1177/0272989x12458724 · Cited by 1,207
Jansen, J. P., Fleurence, R., Devine, B., Itzler, R., Barrett, A., Hawkins, N., Lee, K., Boersma, C., Annemans, L., & Cappelleri, J. C. (2011) · Value in Health · doi:10.1016/j.jval.2011.04.002 · Cited by 1,036
Dias, S., Welton, N. J., Sutton, A. J., & Ades, A. E. (2011) · Bristol Research (University of Bristol) · Cited by 700
Price, M. J., Ades, A., Soldan, K., Welton, N. J., Macleod, J., Simms, I., DeAngelis, D., Turner, K. M., & Horner, P. J. (2016) · Health Technology Assessment · doi:10.3310/hta20220 · Cited by 494
“In a good-fitting model, the summed residual deviances should approximate to the number of data points.(156,157) When comparing models with different numbers of parameters a common approach in classical statistics is the Akaike Information criterion,(158) in which the models are ‘penalised’ by adding the number of par…”
MethodTocolytic therapy for preterm delivery: systematic review and network meta-analysis
Haas, D. M., Caldwell, D. M., Kirkpatrick, P., McIntosh, J. J., & Welton, N. J. (2012) · BMJ · doi:10.1136/bmj.e6226 · Cited by 419
Schoser, B., Stewart, A., Kanters, S., Hamed, A., Jansen, J., Chan, K., Karamouzian, M., & Toscano, A. (2017) · Journal of Neurology · doi:10.1007/s00415-016-8219-8 · Cited by 259
“The DIC provides a measure of model fit that penalizes model in a manner similar to the Akaike Information Criterion (AIC) used in frequentist scenarios [14, 15].”
MethodCooper, N. J., Sutton, A. J., Morris, D., Ades, A. E., & Welton, N. J. (2009) · Statistics in Medicine · doi:10.1002/sim.3594 · Cited by 237
“Under the null hypothesis that the model provides an adequate fit to the data, it is expected that D̄ would have a mean approximately equal to the number of unconstrained data points (this is exact if the data have a Normal likelihood [11]).”
BackgroundUterotonic agents for preventing postpartum haemorrhage: a network meta-analysis
Gallos, I. D., Williams, H. M., Price, M. J., Merriel, A., Gee, H., Lissauer, D., Moorthy, V., Tobias, A., Deeks, J. J., Widmer, M., et al. (2018) · Cochrane Database of Systematic Reviews · doi:10.1002/14651858.cd011689.pub2 · Cited by 228
Network meta-analysis of survival data with fractional polynomials
Jansen, J. P. (2011) · BMC Medical Research Methodology · doi:10.1186/1471-2288-11-61 · Cited by 166
“In a Bayesian framework the Bayesian information criterion (BIC) or deviance information criterion (DIC) can be used [15,16].”
BackgroundMethodPrice, M. J., Ades, A. E., De Angelis, D., Welton, N. J., Macleod, J., Soldan, K., Simms, I., Turner, K., & Horner, P. J. (2013) · American Journal of Epidemiology · doi:10.1093/aje/kws583 · Cited by 164
“To assess goodness of fit, we used the posterior mean residual deviance, whose expected value is approximately equal to the number of data points under the assumption that the model is true (35, 37, 38).”
MethodBayes factors: Prior sensitivity and model generalizability
Liu, C. C., & Aitkin, M. (2008) · Journal of Mathematical Psychology · doi:10.1016/j.jmp.2008.03.002 · Cited by 163
Variational approximations in Bayesian model selection for finite mixture distributions
McGrory, C. A., & Titterington, D. M. (2007) · Computational Statistics & Data Analysis · doi:10.1016/j.csda.2006.07.020 · Cited by 169
“Using the posterior mean deviance as measure of fit was suggested by Dempster (1974) who proposed using the posterior mean of the classical deviance statistic to perform Bayesian model selection.”
MethodUrban land uses, socio-demographic attributes and commuting: A multilevel modeling approach
Antipova, A., Wang, F., & Wilmot, C. G. (2011) · Applied Geography · doi:10.1016/j.apgeog.2011.02.001 · Cited by 152
Bayesian Measures of Model Complexity and Fit
Spiegelhalter, D. J., Best, N. G., Carlin, B. P., & Van Der Linde, A. (2002) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/1467-9868.00353 · Cited by 151
“Burnham and Anderson (1998) (page 244) pointed out that”
BackgroundMethodInfluentialDynamic Latent Trait Models for Multidimensional Longitudinal Data
Dunson, D. B. (2003) · Journal of the American Statistical Association · doi:10.1198/016214503000000387 · Cited by 145
Projections of lung cancer mortality in West Germany: a case study in Bayesian prediction
Knorr‐Held, L. (2001) · Biostatistics · doi:10.1093/biostatistics/2.1.109 · Cited by 138
“Dempster (1997) and Spiegelhalter et al. (1998) have suggested using the posterior distribution of the (binomial) log-likelihood of the data as a measure of fit, equivalent to examining the posterior distribution of the saturated deviance D = I∑ i=1 J∑ j=1 d2i j (8) with the squared deviance…”
MethodCaldwell, D. M., Welton, N. J., & Ades, A. E. (2010) · Journal of Clinical Epidemiology · doi:10.1016/j.jclinepi.2009.08.025 · Cited by 142
“Because the likelihood is normal, the individual contributions to the residual deviance have a c(2) distribution under the null hypothesis of consistency [27].”
BackgroundGelfand, A. E., & Kottas, A. (2002) · Journal of Computational and Graphical Statistics · doi:10.1198/106186002760180518 · Cited by 131
“In this way we can sample the posterior of the log-likelihood, that is, [ = 1 log f ( y i ; G ) j D ] which is ofinterest for model determination,as in Dempster (1997).”
MethodAdes, A. E. (2003) · Statistics in Medicine · doi:10.1002/sim.1566 · Cited by 128
Stevens, J., Simpson, E. L., Harnan, S., Squires, H., Meng, Y., Thomas, S., Michaels, J. A., & Stansby, G. P. (2012) · British journal of surgery · doi:10.1002/bjs.8895 · Cited by 127
Bayesian Graphical Modelling: A Case-Study in Monitoring Health Outcomes
Spiegelhalter, D. J. (1998) · Journal of the Royal Statistical Society Series C: Applied Statistics · doi:10.1111/1467-9876.00101 · Cited by 122
“The posterior distribution of the total deviance can be obtained, although the appropriate means of comparing these between models is still somewhat an open question (Dempster, 1974; Richardson and Green, 1997).”
BackgroundTesting by Betting: A Strategy for Statistical and Scientific Communication
Shafer, G. R. (2021) · Journal of the Royal Statistical Society Series A (Statistics in Society) · doi:10.1111/rssa.12647 · Cited by 304
“A test with a conventional significance level and high power against a very distinct alternative rejects the null hypothesis with a borderline outcome even though the likelihood ratio favours the null (Dempster, 1997 , pp. 249– 250).”
Network meta‐analysis of parametric survival curves
Ouwens, M., Philips, Z., & Jansen, J. P. (2010) · Research Synthesis Methods · doi:10.1002/jrsm.25 · Cited by 119
“The goodness-of-fit of the model predictions to the observed data was measured by calculating the posterior mean residual deviance, D (Dempster, 1997).”
MethodPerlman, M. D., & Wu, L. (1999) · Statistical Science · doi:10.1214/ss/1009212517 · Cited by 114
“Similar examples have been frequently noted Že.g., Dempster, 1997, pages 249 and 250; Berger and Wolpert, 1988, Example 4a, page 8; Aitken, .1991, page 112; Royall, 1997, pages 16 and 17 .”
BackgroundEvaluating novel agent effects in multiple-treatments meta-regression
Salanti, G., Dias, S., Welton, N. J., Ades, A. E., Golfinopoulos, V., Kyrgiou, M., Mauri, D., & Ioannidis, J. P. A. (2010) · Statistics in Medicine · doi:10.1002/sim.4001 · Cited by 111
Cope, S., Zhang, J., Saletan, S., Smiechowski, B., Jansen, J. P., & Schmid, P. (2014) · BMC Medicine · doi:10.1186/1741-7015-12-93 · Cited by 110
“The best fitting first order fractional polynomial model was selected [36] using the deviance information criteria (DIC) [37,38].”
Multilevel IRT using dichotomous and polytomous response data
Fox, J. (2005) · British Journal of Mathematical and Statistical Psychology · doi:10.1348/000711005x38951 · Cited by 113
“Dempster (1997) and Aitkin (1997), considered the posterior distribution of the loglikelihood ratio (LR).”
BackgroundActuarial Modeling with MCMC and BUGs
Scollnik, D. P. M. (2001) · North American Actuarial Journal · doi:10.1080/10920277.2001.10595987 · Cited by 106
“Dempster (1997) suggested that one might examine the posterior distribution of the loglikelihood to assist with model selection, and we will adopt this simple strategy by monitoring the value of NLL and favoring the model with the smallest posterior expected value for this variable.”
MethodCan a significance test be genuinely Bayesian?
Pereira, C. A. D. B., Stern, J. M., & Wechsler, S. (2008) · Bayesian Analysis · doi:10.1214/08-ba303 · Cited by 98
“For additional discussion of alternative p-values see Dempster (1997) and Kempthorne and Folks (1971).”
BackgroundA Bayesian network meta-analysis for binary outcome: how to do it
Greco, T., Landoni, G., Biondi‐Zoccai, G., D’Ascenzo, F., & Zangrillo, A. (2016) · Statistical Methods in Medical Research · doi:10.1177/0962280213500185 · Cited by 107
Jansen, J. P., & Cope, S. (2012) · BMC Medical Research Methodology · doi:10.1186/1471-2288-12-152 · Cited by 92
“The deviance information criterion (DIC) was used to compare the goodness-of-fit of the different models [27,28].”
MethodJohnson, T. R. (2003) · Psychometrika · doi:10.1007/bf02295612 · Cited by 106
“The DIC is based on the notion of "Bayesian deviance" ( Dempster, 1974, 1997 ) which is defined for the models discussed in this paper as”
MethodCombining MCMC with ‘sequential’ PKPD modelling
Lunn, D. J., Best, N., Spiegelhalter, D. J., Graham, G., & Neuenschwander, B. E. (2009) · Journal of Pharmacokinetics and Pharmacodynamics · doi:10.1007/s10928-008-9109-1 · Cited by 90
“This can be taken as a measure of model fit [24], with lower numbers indicating a better fit to the data.”
Treatments for bulimia nervosa: a network meta-analysis
Slade, E., Keeney, E., Mavranezouli, I., Dias, S., Fou, L., Stockton, S., Saxon, L., Waller, G., Turner, H., Serpell, L., et al. (2018) · Psychological Medicine · doi:10.1017/s0033291718001071 · Cited by 79
A Conditional Joint Modeling Approach for Locally Dependent Item Responses and Response Times
Meng, X., Tao, J., & Chang, H. (2015) · Journal of Educational Measurement · doi:10.1111/jedm.12060 · Cited by 74
Giovane, C. D., Vacchi, L., Mavridis, D., Filippini, G., & Salanti, G. (2013) · Statistics in Medicine · doi:10.1002/sim.5512 · Cited by 65
Comparative efficacy of long-acting bronchodilators for COPD - a network meta-analysis
Cope, S. F., Donohue, J. F., Jansen, J. P., Kraemer, M., Capkun-Niggli, G., Baldwin, M., Buckley, F., Ellis, A. G., & Jones, P. (2013) · Respiratory Research · doi:10.1186/1465-9921-14-100 · Cited by 68
“The deviance information criterion was used to compare the models, which provides a measure of model fit that penalizes model complexity accordingly [23].”
MethodWelton, N. J., Cooper, N. J., Ades, A. E., Lu, G., & Sutton, A. J. (2008) · Statistics in Medicine · doi:10.1002/sim.3377 · Cited by 60
Price, M. J., Ades, A. E., Welton, N. J., Simms, I., Macleod, J., & Horner, P. (2016) · The Journal of Infectious Diseases · doi:10.1093/infdis/jiw178 · Cited by 61
“To assess goodness of fit, we use the posterior mean residual deviance, whose expected value approximates the number of data points under the assumption that the model is true [25, 26].”
Methodda Silveira, N. J. F., Varuzza, L., Machado‐Lima, A., Lauretto, M. D. S., Pinheiro, D. G., Rodrigues, R. V., Severino, P., Nóbrega, F. G., Head and Neck Genome Project GENCAPO, Silva, W. A., et al. (2008) · BMC Medical Genomics · doi:10.1186/1755-8794-1-56 · Cited by 57
“For this task, we decided to use the method developed by Dempster (1997) [16].”
MethodPrice, M. J., Ades, A. E., Angelis, D. D., Welton, N. J., Macleod, J., Soldan, K., Turner, K., Simms, I., & Horner, P. (2012) · Statistics in Medicine · doi:10.1002/sim.5603 · Cited by 58
“In a well-fitting model, the posterior mean residual deviance should be approximately the same as the number of data points [41, 42].”
BackgroundFin whale movements in the Gulf of California, Mexico, from satellite telemetry
Jiménez-López, M. E., Palacios, D. M., Jaramillo‐Legorreta, A. M., Jorge Urbán, R., & Mate, B. R. (2019) · PLoS ONE · doi:10.1371/journal.pone.0209324 · Cited by 61
There Is Individualized Treatment. Why Not Individualized Inference?
Liu, K., & Meng, X. (2016) · Annual Review of Statistics and Its Application · doi:10.1146/annurev-statistics-010814-020310 · Cited by 59
“This example comes from Dempster (1997) and is extended using ideas in Berger (2003).”
MethodWang, F., Luo, L. F., & McLafferty, S. (2009) · International Journal of Public Policy · doi:10.1504/ijpp.2010.030606 · Cited by 53
Dakin, H. A., Welton, N. J., Ades, A. E., Collins, S., Orme, M., & Kelly, S. (2011) · Statistics in Medicine · doi:10.1002/sim.4284 · Cited by 53
“55 SEs; by contrast, residual deviance for a well-fitting model of 742 datapoints should be around 742 [46].”
BackgroundMethodLindsey, J. K. (1999) · Document Server@UHasselt (UHasselt) · doi:10.1111/j.1467-9884.1999.tb09211.x · Cited by 51
Elborn, J. S., Vataire, A. L., Fukushima, A., Aballéa, S., Khemiri, A., Moore, C., Medić, G., & Hemels, M. E. H. (2016) · Clinical Therapeutics · doi:10.1016/j.clinthera.2016.08.014 · Cited by 56
“The deviance information criterion was used to compare the models, which provides a measure of model fit that penalizes model complexity accordingly.(17) All models were based on the NICE Decision Support Unit”
MethodHolsclaw, T., Greene, A. M., Robertson, A. W., & Smyth, P. (2017) · The Annals of Applied Statistics · doi:10.1214/16-aoas1009 · Cited by 65
“Bayes factors (BF) are not considered because the model has non-informative priors on many of the parameters (Kass and Raftery, 1995; Dempster, 1997).”
BackgroundMethodKarabis, A., Lindner, L., Mocarski, M., Huisman, & Greening (2013) · International Journal of COPD · doi:10.2147/copd.s48967 · Cited by 44
“15 For all continuous outcomes, a generalized linear model with identity link and a normal likelihood distribution was used, 16,17 while a logit link with binomial likelihood distribution was used for dichotomous outcomes.”
MethodCope, S. F., Ayers, D., Zhang, J., Batt, K., & Jansen, J. P. (2019) · BMC Medical Research Methodology · doi:10.1186/s12874-019-0823-8 · Cited by 51
Edwards, S. J., & Smith, C. J. (2009) · Clinical Therapeutics · doi:10.1016/j.clinthera.2009.07.004 · Cited by 46
Turner, N. L., Dias, S., Ades, A. E., & Welton, N. J. (2015) · Statistics in Medicine · doi:10.1002/sim.6475 · Cited by 42
“The posterior mean residual deviance [29, 31] suggests that a random effects model is a better fit to the data than a fixed effects model in all cases (Table III).”
BackgroundBayesian point null hypothesis testing via the posterior likelihood ratio
Aitkin, M. A., Boys, R. J., & Chadwick, T. J. (2005) · Statistics and Computing · doi:10.1007/s11222-005-1310-0 · Cited by 41
Edwards, S. J., Lind, T., Lundell, L. R., & Das, R. (2009) · Alimentary Pharmacology & Therapeutics · doi:10.1111/j.1365-2036.2009.04077.x · Cited by 44
Estimation of the limit of detection using information theory measures
Fonollosa, J., Vergara, A., Huerta, R., & Marco, S. (2013) · Analytica Chimica Acta · doi:10.1016/j.aca.2013.10.030 · Cited by 40
Bayesian analysis of simulation-based models
Turner, B. M., Sederberg, P. B., & McClelland, J. L. (2014) · Journal of Mathematical Psychology · doi:10.1016/j.jmp.2014.10.001 · Cited by 51
“The effective number of parameters is based on the difference between the expected deviance and an estimate of the deviance at the most likely value of the parameters (Dempster, 1997).”
BackgroundCope, S. F., Ouwens, M., Jansen, J. P., & Schmid, P. (2013) · Value in Health · doi:10.1016/j.jval.2012.10.019 · Cited by 42
Bayesian evidence synthesis for a transmission dynamic model for HIV among men who have sex with men
Presanis, A. M., Angelis, D. D., Goubar, A., Gill, O. N., & Ades, A. E. (2011) · Biostatistics · doi:10.1093/biostatistics/kxr006 · Cited by 37
“To assess absolute goodness of fit, we compare the posterior mean deviance D̄ = Eθ |y {D(θ )} =∫ D(θ , y)p(θ |y)dθ with the number of observations n (Dempster, 1997; Spiegelhalter and others, 2002).”
MethodConflict Diagnostics in Directed Acyclic Graphs, with Applications in Bayesian Evidence Synthesis
Presanis, A. M., Ohlssen, D. I., Spiegelhalter, D. J., & Angelis, D. D. (2013) · Statistical Science · doi:10.1214/13-sts426 · Cited by 54
“One strategy may be to employ a diagnostic such as comparison of posterior mean deviance to the number of data items (Dempster (1997); Spiegelhal-ter et al. (2002)) to detect lack of fit to particular items, which may then be an indicator of con-flicting evidence (Ades and Cliffe (2002); Presanis et…”
MethodPericchi, L. R., & Pereira, C. A. D. B. (2016) · Brazilian Journal of Probability and Statistics · doi:10.1214/14-bjps257 · Cited by 38
Fouskakis, D., Ntzoufras, I., & Draper, D. (2009) · Cited by 39
O’Malley, A. J., Normand, S. T., & Kuntz, R. E. (2002) · Statistics in Medicine · doi:10.1002/sim.1337 · Cited by 36
Bayesian hierarchical modeling of Pacific geoduck growth increment data and climate indices
Helser, T. E., Lai, H.-L., & Black, B. A. (2012) · Ecological Modelling · doi:10.1016/j.ecolmodel.2012.08.024 · Cited by 42
Bayesian Modeling of the Level and Duration of Fertility in the Menstrual Cycle
Dunson, D. B. (2001) · Biometrics · doi:10.1111/j.0006-341x.2001.01067.x · Cited by 36
“The link functions in expression (8) were chosen as the best fitting forms based on a comparison of the predictive log likelihoods (Dempster, 1974) under probit, logistic, complementary -log-log, and exponential choices.”
MethodInhibition in Multiclass Classification
Huerta, R., Vembu, S., Amigó, J. M., Nowotny, T., & Elkan, C. (2012) · Neural Computation · doi:10.1162/neco_a_00321 · Cited by 36
“For simplicity, we prefer to solve the problem of inhibitory classifiers in the framework of Diettrich and Bakiri (1995). The extension proposed by Allwein et al.”
MethodConflict Diagnostics in Directed Acyclic Graphs, with Applications in Bayesian Evidence Synthesis
Presanis, A. M., Ohlssen, D. I., Spiegelhalter, D. J., & Angelis, D. D. (2013) · Apollo (University of Cambridge) · doi:10.17863/cam.34842 · Cited by 34
Using Landsat Imagery and Census Data for Urban Population Density Modeling in Port-au-Prince, Haiti
Joseph, M., Wang, L., & Wang, F. (2012) · GIScience & Remote Sensing · doi:10.2747/1548-1603.49.2.228 · Cited by 38
“The AIC is a natural way to compare complex models with prior distributions, in that it is based on the posterior distribution of the log-likelihood, following the Bayesian model framework built by Dempster (1974).”
MethodBayesian models for spatial incidence: a case study of suicide using the BUGS program
Congdon, P. D. (1997) · Health & Place · doi:10.1016/s1353-8292(97)00017-8 · Cited by 31
Rose, M. S., Gillis, A. M., & Sheldon, R. S. (1999) · Statistics in Medicine · doi:10.1002/(sici)1097-0258(19990130)18:2<139::aid-sim9>3.0.co;2-x · Cited by 29
RELATIONSHIP BETWEEN BAYESIAN AND FREQUENTIST SIGNIFICANCE INDICES
Diniz, M. A., Pereira, C. A. D. B., Polpo, A., Stern, J. M., & Wechsler, S. (2012) · International Journal for Uncertainty Quantification · doi:10.1615/int.j.uncertaintyquantification.v2.i2.60 · Cited by 31
“This subject is discussed by [16] and [17].”
BackgroundAitkin, M. A. (1997) · Statistics and Computing · doi:10.1023/a:1018550505678 · Cited by 29
“In a conference paper (reprinted on pp. 247±252), Dempster (1974) considered the problem of testing a simple point null hypothesis against a composite alternative.”
Network meta‐analysis of longitudinal data using fractional polynomials
Jansen, J. P., Vieira, M.-C., & Cope, S. F. (2015) · Statistics in Medicine · doi:10.1002/sim.6492 · Cited by 32
“In a Bayesian framework, the Bayesian information criterion or deviance information criterion (DIC) can be used [29, 30].”
MethodMadan, J. J., Chen, Y., Aveyard, P. N., Wang, D., Yahaya, I., Munafo, M. R., Bauld, L., & Welton, N. J. (2013) · Journal of the Royal Statistical Society Series A (Statistics in Society) · doi:10.1111/rssa.12018 · Cited by 28
“Each data point should contribute approximately 1 to D̄ (Dempster, 1997; Spiegelhalter et al., 2002; Welton et al., 2012), so that we can compare D̄ with the number of independent data points to obtain an approximate check of the fit of a single given model.”
Backgroundde Kleer, K., Luszcz‐Cook, S., de Pater, I., Ádámkovics, M., & Hammel, H. B. (2015) · Icarus · doi:10.1016/j.icarus.2015.04.021 · Cited by 28
“DðhÞ ¼ 2lnLðyjhÞ þ 2lnf ðyÞ; ð4Þ where LðyjhÞ is the likelihood function for the data y given the parameter set h and f ðyÞ is a function of the data alone, and therefore cancels when considering the difference in deviance between models for the same dataset (Dempster, 1974).”
Dequen, P., Lorigan, P. C., Jansen, J. P., van Baardewijk, M., Ouwens, M., & Kotapati, S. (2012) · The Oncologist · doi:10.1634/theoncologist.2011-0427 · Cited by 29
“The goodness of fit of the model predictions to the observed data was measured by calculating the posterior mean residual deviance [30].”
MethodFouskakis, D., Ntzoufras, I., & Draper, D. (2009) · The Annals of Applied Statistics · doi:10.1214/08-aoas207 · Cited by 39
Ye, J., Hu, Y., Chen, X., Chang, C., & Li, K. (2023) · Nutrients · doi:10.3390/nu15122772 · Cited by 31
“If the difference in DIC was less than 5 points, consistency was considered good and consistency modeling was used [34].”
MethodPrevost, T. C., Presanis, A. M., Taylor, A., Goldberg, D. J., Hutchinson, S. J., & Angelis, D. D. (2015) · Addiction · doi:10.1111/add.12948 · Cited by 31
“D̂ is the deviance evaluated at the maximum likelihood result, D is the posterior mean deviance, pD is the number of parameters and DIC is the deviance information criterion [23,24].”
BackgroundCaldwell, D. M., Welton, N. J., Dias, S., & Ades, A. (2012) · Research Synthesis Methods · doi:10.1002/jrsm.1040 · Cited by 27
“If the model fits well, we would expect the posterior mean of the summed deviance contributions to be close to the number of data points, in this case 79 (Dempster, 1997).”
MethodTwo-pronged Strategy for Using DIC to Compare Selection Models with Non-Ignorable Missing Responses
Best, N., Mason, A. J., & Richardson, S. T. (2012) · Bayesian Analysis · doi:10.1214/12-ba704 · Cited by 25
“In Bayesian statistics deviance can be summarised in different ways, with the posterior mean of the deviance, D(ϕ) = E{D(ϕ)|y}, suggested as a sensible Bayesian measure of fit (Dempster, 1973) (reprinted as Dempster (1997)), though this is not penalised for model complexity.”
BackgroundBayesian predictive model comparison via parallel sampling
Congdon, P. D. (2004) · Computational Statistics & Data Analysis · doi:10.1016/j.csda.2004.03.016 · Cited by 24
A Genetic Analysis of Mortality in Pigs
Varona, L., & Sørensen, D. (2009) · Genetics · doi:10.1534/genetics.109.110759 · Cited by 25
“The tables report also the loglikelihood, ln f y jh; uð Þ, averaged over the MCMC replicates, as a measure of model fit (e.g., Dempster 1997).”
MethodPosterior distributions for likelihood ratios in forensic science
van den Hout, A., & Alberink, I. B. (2016) · Science & Justice · doi:10.1016/j.scijus.2016.06.011 · Cited by 24
Bayes Estimate and Inference for Entropy and Information Index of Fit
Mazzuchi, T. A., Soofi, E. S., & Soyer, R. (2008) · Econometric Reviews · doi:10.1080/07474930801960311 · Cited by 23
“Spigelhalter et al. (2002) developed a generalization of the AIC for hierarchical Bayesian models using the posterior deviance concept proposed by Dempster (1974) .”
MethodCope, S. F., Capkun-Niggli, G., Gale, R., Jardim, J. R., & Jansen, J. P. (2011) · International Journal of COPD · doi:10.2147/copd.s18759 · Cited by 22
“11 With a network meta-analysis, randomization only holds within a trial and not across trials.”
Bayesian model comparison and model averaging for small-area estimation
Aitkin, M., Liu, C. C., & Chadwick, T. (2009) · The Annals of Applied Statistics · doi:10.1214/08-aoas205 · Cited by 32
“We set out a general methodology for this purpose, following Dempster (1974, 1997), Aitkin (1997) and Aitkin, Boys and Chadwick (2005); see also Fox (2005).”
MethodInfluentialLogicist statistics. I. Models and modeling
Dempster, A. P. (1998) · Statistical Science · doi:10.1214/ss/1028905887 · Cited by 26
“Dempster, A. P. (1971). Model searching and estimation in the”
InfluentialSelf-citationStatistical Analysis of Skin Tumor Data from Tg.AC Mouse Bioassays
Dunson, D. B. (2000) · Toxicological Sciences · doi:10.1093/toxsci/55.2.293 · Cited by 21
“We used the predictive log-likelihood approach of Dempster (1974), as described in Karim and Zeger (1992), to further”
MethodInfluentialLebwohl, M., Carvalho, A. L., Asahina, A., Zhang, J., Fazeli, M. S., Kasireddy, E., Serafini, P. R., Ferro, T. J., Gogineni, R., & Thaçi, D. (2025) · Dermatology and Therapy · doi:10.1007/s13555-025-01423-0 · Cited by 25
“Model fit was assessed using the deviance information criterion (DIC), a measure of model goodness-of-fit which is penalized for model complexity since a more complex model will result in better fit at the expense of parsimony [16].”
Upadhyay, S. K., Gupta, A., & Dey, D. K. (2012) · Sankhya B · doi:10.1007/s13571-012-0041-4 · Cited by 20
The Language of Betting as a Strategy for Statistical and Scientific CommunicationPreprint
Shafer, G. R. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1903.06991 · Cited by 40
“An article on the topic by Dempster, first published with discussion in the proceedings of a 1973 conference, was reprinted in 1997 along with further discussion [14].”
BackgroundGong, X., He, Z., Li, P., & Zhu, N. (2014) · Discrete Dynamics in Nature and Society · doi:10.1155/2014/964654 · Cited by 19
“Dempster [41] considered that posterior distribution inspecting the classical deviation can employBayesianmodel, that is:”
BackgroundKanters, S., Ball, G., Kahl, B. S., Wiesinger, A., Limbrick‐Oldfield, E. H., Sudhindra, A., Snider, J. T., & Patel, A. R. (2023) · BMC Cancer · doi:10.1186/s12885-023-10546-6 · Cited by 23
“For Bayesian analyses, the deviance information criterion (DIC) was used to compare the goodness-of-fit of competing survival models [30].”
MethodCope, S. F., Capkun-Niggli, G., Gale, R., Lassen, C., Owen, R. J., Ouwens, M., Bergman, G., & Jansen, J. P. (2012) · Value in Health · doi:10.1016/j.jval.2012.01.009 · Cited by 18
BAYESIAN MODEL COMPARISON AND MODEL AVERAGING FOR SMALL-AREA ESTIMATION 1Preprint
Aitkin, M. A., Liu, C. C., & Chadwick, T. J. (2006) · Cited by 19
Information importance of predictors: Concept, measures, Bayesian inference, and applications
Retzer, J. J., Soofi, E. S., & Soyer, R. (2008) · Computational Statistics & Data Analysis · doi:10.1016/j.csda.2008.03.010 · Cited by 21
Bayes Analysis and Comparison of Accelerated Weibull and Accelerated Birnbaum–Saunders Models
Upadhyay, S. K., & Mukherjee, B. (2009) · Communication in Statistics- Theory and Methods · doi:10.1080/03610920802663323 · Cited by 18
“The formulation of DIC is based on examining the posterior distribution of the classical deviance (see Dempster, 1974 ), which can be defined as where log h ( t ) denotes a fully specified standardizing term and is a function of data alone.”
MethodGeostatistical modelling for spatial interaction data with application to postal service performance
Banerjee, S., Gelfand, A. E., & Polasek, W. (2000) · Journal of Statistical Planning and Inference · doi:10.1016/s0378-3758(00)00111-7 · Cited by 18
Koskinen, J. H. (2008) · Research Explorer (The University of Manchester) · Cited by 25
Metadata
- Authors
- Dempster, A. P.
- Affiliations
- Harvard University (US)
- Corresponding author
- Dempster, A. P.
- Venue
- Statistics and Computing
- Published
- 1997-12
- Volume
- 7
- Issue
- 4
- Pages
- 247-252
- ISSN
- 0960-3174 (ISSN-L)
- 1573-1375
- OpenAlex
- W1569625389
- Cited by
- 220
- References
- 5
- Citations by year
- 2026: 4 · 2025: 15 · 2024: 11 · 2023: 4 · 2022: 8 · 2021: 3 · 2020: 2 · 2019: 5 · 2018: 4 · 2017: 4 · 2016: 11 · 2015: 12 · 2014: 6 · 2013: 17 · 2012: 21
- FWCI
- 5.61
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- www.springernature.com/gp/researchers/text-and-data-mining (tdm)
- www.springernature.com/gp/researchers/text-and-data-mining (vor)
- Locations
- Statistics and Computing (publishedVersion)
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Physical Sciences › Mathematics › Statistics and Probability › Probability and Statistical Research
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Inference
- Keywords
- likelihood ratios, significance testing, Bayesian hypothesis testing
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.