Tutorial on maximum likelihood estimation

Myung, In Jae

doi:10.1016/s0022-2496(02)00028-7

Cited by 1,424 publications

(866 citation statements)

References 14 publications

Supporting

Mentioning

835

Contrasting

Unclassified

Order By: Relevance

“…The authors acknowledged that the gold standard approach to estimate NTCP parameters is the use of individual dosimetric data/clinical outcome. MLE method based parameter estimates are also more precise as compared to weighted least square method [29].…”

Section: Discussionmentioning

confidence: 99%

Normal tissue complication probability (NTCP) parameters for breast fibrosis: Pooled results from two randomised trials

Mukesh

Harris

Collette

et al. 2013

Radiotherapy and Oncology

View full text Add to dashboard Cite

a b s t r a c tIntroduction: The dose-volume effect of radiation therapy on breast tissue is poorly understood. We estimate NTCP parameters for breast fibrosis after external beam radiotherapy. Materials and methods: We pooled individual patient data of 5856 patients from 2 trials including whole breast irradiation followed with or without a boost. A two-compartment dose volume histogram model was used with boost volume as the first compartment and the remaining breast volume as second compartment. Results from START-pilot trial (n = 1410) were used to test the predicted models. Results: 26.8% patients in the Cambridge trial (5 years) and 20.7% patients in the EORTC trial (10 years) developed moderate-severe breast fibrosis. The best fit NTCP parameters were BEUD 3 (50) = 136.4 Gy, c50 = 0.9 and n = 0.011 for the Niemierko model and BEUD 3 (50) = 132 Gy, m = 0.35 and n = 0.012 for the Lyman Kutcher Burman model. The observed rates of fibrosis in the START-pilot trial agreed well with the predicted rates. Conclusions: This large multi-centre pooled study suggests that the effect of volume parameter is small and the maximum RT dose is the most important parameter to influence breast fibrosis. A small value of volume parameter 'n' does not fit with the hypothesis that breast tissue is a parallel organ. However, this may reflect limitations in our current scoring system of fibrosis.

show abstract

Section: Discussionmentioning

confidence: 99%

Normal tissue complication probability (NTCP) parameters for breast fibrosis: Pooled results from two randomised trials

Mukesh

Harris

Collette

et al. 2013

Radiotherapy and Oncology

View full text Add to dashboard Cite

show abstract

“…This approach uses the model to assign a probability to the observed data, and finding the parameter values that maximize this probability. The value of this maximized probability is called the maximum likelihood (ML; see Myung 2003) The ML method permits the specification of an appropriate error distribution, thereby avoiding the skewness problem. For models that predict that p(C) data arise as an average proportion of success across a series of independent (Bernoulli) trials, the desired error distribution is the binomial.…”

Section: Methodological Issuesmentioning

confidence: 99%

Assessing the distinguishability of models and the informativeness of data

Navarro

Pitt

Myung

2004

Cognitive Psychology

Self Cite

View full text Add to dashboard Cite

“…The central tendency of a posterior distribution is often summarized by its mean, median, or mode. Note that with a uniform prior, the mode of a posterior distribution coincides with the classical maximum likelihood estimate or MLE,ĥ ¼ s=n ¼ 0:9 (Myung, 2003). The spread of a posterior distribution is most easily captured by a Bayesian x% confidence interval that extends from the ðx=2Þth to the ð100 À x=2Þth percentile of the posterior distribution.…”

Section: Bayesian Parameter Estimationmentioning

confidence: 98%

Bayesian hypothesis testing for psychologists: A tutorial on the Savage–Dickey method

Wagenmakers

Lodewyckx

Kuriyal

et al. 2010

Cognitive Psychology

652

642

View full text Add to dashboard Cite

a b s t r a c tIn the field of cognitive psychology, the p-value hypothesis test has established a stranglehold on statistical reporting. This is unfortunate, as the p-value provides at best a rough estimate of the evidence that the data provide for the presence of an experimental effect. An alternative and arguably more appropriate measure of evidence is conveyed by a Bayesian hypothesis test, which prefers the model with the highest average likelihood. One of the main problems with this Bayesian hypothesis test, however, is that it often requires relatively sophisticated numerical methods for its computation. Here we draw attention to the Savage-Dickey density ratio method, a method that can be used to compute the result of a Bayesian hypothesis test for nested models and under certain plausible restrictions on the parameter priors. Practical examples demonstrate the method's validity, generality, and flexibility.

show abstract

Tutorial on maximum likelihood estimation

Cited by 1,424 publications

References 14 publications

Normal tissue complication probability (NTCP) parameters for breast fibrosis: Pooled results from two randomised trials

Normal tissue complication probability (NTCP) parameters for breast fibrosis: Pooled results from two randomised trials

Assessing the distinguishability of models and the informativeness of data

Bayesian hypothesis testing for psychologists: A tutorial on the Savage–Dickey method

Contact Info

Product

Resources

About