# uncertainty

**URL:** https://discourse.datamethods.org/c/stat/uncertain/8.md

[Latest](https://discourse.datamethods.org/latest.md) · [Categories](https://discourse.datamethods.org/categories.md) · [Tags](https://discourse.datamethods.org/tags.md)

---

## [About the uncertainty category](https://discourse.datamethods.org/t/about-the-uncertainty-category/21)

<div class="topic-metadata">

**Author:** [@f2harrell](https://discourse.datamethods.org/u/f2harrell)\
**Replies:** 0

</div>

Quantifying uncertainty, displaying uncertainty, estimation of uncertainty, incorporation of uncertainty into decision making, etc. This includes but is not limited to confidence intervals, standard errors, Bayesian cre…

---

## [Calculating the variance of mean event count in a vaccine study re-analysis](https://discourse.datamethods.org/t/calculating-the-variance-of-mean-event-count-in-a-vaccine-study-re-analysis/28772)

<div class="topic-metadata">

**Author:** [@tamas.ferenci](https://discourse.datamethods.org/u/tamas.ferenci)\
**Replies:** 0\
**Last updated:** [June 2, 2026, 6:49pm UTC](https://discourse.datamethods.org/t/calculating-the-variance-of-mean-event-count-in-a-vaccine-study-re-analysis/28772 "2026-06-02T18:49:09Z")

</div>

In 2022, Fraiman et al published a controversial paper about the adverse events of COVID vaccines. It was controversial primarily because many felt that it not only has flaws, but that its mistakes are not random, rather…

---

## [Confidence intervals for right-censored outcome with no events](https://discourse.datamethods.org/t/confidence-intervals-for-right-censored-outcome-with-no-events/28731)

<div class="topic-metadata">

**Author:** [@nsrath](https://discourse.datamethods.org/u/nsrath)\
**Replies:** 2\
**Last updated:** [May 4, 2026, 10:30pm UTC](https://discourse.datamethods.org/t/confidence-intervals-for-right-censored-outcome-with-no-events/28731 "2026-05-04T22:30:29Z")

</div>

In the context of a variable with a binomial distribution, there are methods like the Wilson score interval that produce sensible confidence intervals even when no events are observed For an analysis I’ve been working …

---

## [Good ways to verify coverage of CIs for probabilities in variational bayes binary classification?](https://discourse.datamethods.org/t/good-ways-to-verify-coverage-of-cis-for-probabilities-in-variational-bayes-binary-classification/28450)

<div class="topic-metadata">

**Author:** [@ludvighult](https://discourse.datamethods.org/u/ludvighult)\
**Replies:** 3\
**Last updated:** [October 9, 2025, 7:57am UTC](https://discourse.datamethods.org/t/good-ways-to-verify-coverage-of-cis-for-probabilities-in-variational-bayes-binary-classification/28450 "2025-10-09T07:57:07Z")

</div>

In short: a bayesian binary predictor can provide a posterior on the predicted probability, providing showing a point prediction (e.g. this patient has 3-22% risk) instead of just providing point predictions (this patien…

---

## [Computing Different Standard Errors and Confidence Intervals](https://discourse.datamethods.org/t/computing-different-standard-errors-and-confidence-intervals/28010)

<div class="topic-metadata">

**Author:** [@rafdoodle1](https://discourse.datamethods.org/u/rafdoodle1)\
**Replies:** 15\
**Last updated:** [May 12, 2025, 5:59pm UTC](https://discourse.datamethods.org/t/computing-different-standard-errors-and-confidence-intervals/28010 "2025-05-12T17:59:06Z")

</div>

Hello Dr. Harrell, Hope you are doing well. To compute confidence intervals for Shap values (or even relative explained variation for that matter), I would need their standard errors if I can recall. The iml R package g…

---

## [Bootstrap ranking for differntially expressed proteins in linear mixed model vs t-tests](https://discourse.datamethods.org/t/bootstrap-ranking-for-differntially-expressed-proteins-in-linear-mixed-model-vs-t-tests/27906)

<div class="topic-metadata">

**Author:** [@Natan\_Nagar](https://discourse.datamethods.org/u/Natan_Nagar)\
**Replies:** 3\
**Last updated:** [April 23, 2025, 2:48pm UTC](https://discourse.datamethods.org/t/bootstrap-ranking-for-differntially-expressed-proteins-in-linear-mixed-model-vs-t-tests/27906 "2025-04-23T14:48:32Z")

</div>

Sorry if this is an elementary question. I was tasked with analyzing repeated measures of protein expression (baseline and 6-month follow-up; 2 time points so far) across two treatment groups (44 subjects, 357 features).…

---

## [Between Subjects Variance and Within Subjects Variance in ICC](https://discourse.datamethods.org/t/between-subjects-variance-and-within-subjects-variance-in-icc/6031)

<div class="topic-metadata">

**Author:** [@Sudhi\_Upadhyaya](https://discourse.datamethods.org/u/Sudhi_Upadhyaya)\
**Replies:** 2\
**Last updated:** [September 25, 2022, 10:15pm UTC](https://discourse.datamethods.org/t/between-subjects-variance-and-within-subjects-variance-in-icc/6031 "2022-09-25T22:15:32Z")

</div>

Sorry this may be a naïve question. Generally ICC is considered an indicator of reproducibility. Higher ICC is interpreted as good reproducibility , lower ICC not so much. ICC as such is a ratio of between subjects vari…

---

## [Uncertainty of individual patient pharmacokinetic parameters](https://discourse.datamethods.org/t/uncertainty-of-individual-patient-pharmacokinetic-parameters/5714)

<div class="topic-metadata">

**Author:** [@eliast](https://discourse.datamethods.org/u/eliast)\
**Replies:** 2\
**Last updated:** [May 31, 2022, 4:19pm UTC](https://discourse.datamethods.org/t/uncertainty-of-individual-patient-pharmacokinetic-parameters/5714 "2022-05-31T16:19:39Z")

</div>

In daily practice we are often asked to estimate the ellimenation rate constant of a drug in a patient in order to choose a suitable dosing regimen or calculate 24- hour area under the concentration time curve. Usually ,…

---

## [Power calculation for limits of agreement of imaging algorithm vs expert, to be applied with a lay-human in the loop](https://discourse.datamethods.org/t/power-calculation-for-limits-of-agreement-of-imaging-algorithm-vs-expert-to-be-applied-with-a-lay-human-in-the-loop/5321)

<div class="topic-metadata">

**Author:** [@tkaczye](https://discourse.datamethods.org/u/tkaczye)\
**Replies:** 3\
**Last updated:** [February 14, 2022, 12:52pm UTC](https://discourse.datamethods.org/t/power-calculation-for-limits-of-agreement-of-imaging-algorithm-vs-expert-to-be-applied-with-a-lay-human-in-the-loop/5321 "2022-02-14T12:52:35Z")

</div>

Would really appreciate anyone’s advice on how to fairly estimate performance of a measurement algorithm when a human-in-the-loop will determine the times it is applied. Specifics: We have preliminary data for a deep l…

---

## [Determine power to detect a misclassification event in rule-out pathways for myocardial infarction](https://discourse.datamethods.org/t/determine-power-to-detect-a-misclassification-event-in-rule-out-pathways-for-myocardial-infarction/5258)

<div class="topic-metadata">

**Author:** [@tomkaier](https://discourse.datamethods.org/u/tomkaier)\
**Replies:** 10\
**Last updated:** [January 15, 2022, 11:32pm UTC](https://discourse.datamethods.org/t/determine-power-to-detect-a-misclassification-event-in-rule-out-pathways-for-myocardial-infarction/5258 "2022-01-15T23:32:32Z")

</div>

The question concerns work we have recently conducted, looking at risk prediction using a serum biomarker in a study population presenting with chest pain to emergency departments. In clinical medicine, biomarkers such a…

---

## [Early use of tocilizumab and steroids does not impact expansion or persistence of CAR T cells in B-ALL?](https://discourse.datamethods.org/t/early-use-of-tocilizumab-and-steroids-does-not-impact-expansion-or-persistence-of-car-t-cells-in-b-all/4141)

<div class="topic-metadata">

**Author:** [@karlamoPA](https://discourse.datamethods.org/u/karlamoPA)\
**Replies:** 2\
**Last updated:** [February 17, 2021, 2:58pm UTC](https://discourse.datamethods.org/t/early-use-of-tocilizumab-and-steroids-does-not-impact-expansion-or-persistence-of-car-t-cells-in-b-all/4141 "2021-02-17T14:58:50Z")

</div>

Seeking comment on the methods used in this study, where the authors conclude as shown in bulleted key points below. As a patient advocate I am not qualified to critique the methods or conclusion but have read other art…

---

## [Visualization for Benefit of COVID-19 Testing for Decreasing Provider Risk During Intubation](https://discourse.datamethods.org/t/visualization-for-benefit-of-covid-19-testing-for-decreasing-provider-risk-during-intubation/3164)

<div class="topic-metadata">

**Author:** [@dr.vancleve](https://discourse.datamethods.org/u/dr.vancleve)\
**Replies:** 1\
**Last updated:** [April 2, 2020, 12:33pm UTC](https://discourse.datamethods.org/t/visualization-for-benefit-of-covid-19-testing-for-decreasing-provider-risk-during-intubation/3164 "2020-04-02T12:33:37Z")

</div>

Datamethods Colleagues - I am an anesthesiologist, and our largest professional society recently recommended that healthcare workers trained to place breathing tubes consider wearing a specialized, increasingly rare res…

---

## ["Impressive" or "anecdotal"? Hydroxychloroquine + Azithromycin vs COVID-19 preprint](https://discourse.datamethods.org/t/impressive-or-anecdotal-hydroxychloroquine-azithromycin-vs-covid-19-preprint/3118)

<div class="topic-metadata">

**Author:** [@davidcnorrismd](https://discourse.datamethods.org/u/davidcnorrismd)\
**Replies:** 4\
**Last updated:** [March 22, 2020, 2:27pm UTC](https://discourse.datamethods.org/t/impressive-or-anecdotal-hydroxychloroquine-azithromycin-vs-covid-19-preprint/3118 "2020-03-22T14:27:00Z")

</div>

Discussed yesterday at both the highest and lowest levels … https://www.nytimes.com/video/us/politics/100000007046134/trump-fauci.html … this study \[1\] apparently dichotomized a PCR cycle threshold C\_t such that C\_t \\ge…

---

## [Confidence band for the ROC curve](https://discourse.datamethods.org/t/confidence-band-for-the-roc-curve/2969)

<div class="topic-metadata">

**Author:** [@jlevy13](https://discourse.datamethods.org/u/jlevy13)\
**Replies:** 2\
**Last updated:** [February 10, 2020, 9:34am UTC](https://discourse.datamethods.org/t/confidence-band-for-the-roc-curve/2969 "2020-02-10T09:34:19Z")

</div>

How would you construct a confidence band for the ROC curve? How would you construct a confidence band for the ROC curve? I do not mean a confidence interval for the AUC or for other parameters that can be derived from…

---

## [Informative priors to reduce prediction uncertainty](https://discourse.datamethods.org/t/informative-priors-to-reduce-prediction-uncertainty/2957)

<div class="topic-metadata">

**Author:** [@Sreenath\_Madathil](https://discourse.datamethods.org/u/Sreenath_Madathil)\
**Replies:** 1\
**Last updated:** [February 6, 2020, 3:12am UTC](https://discourse.datamethods.org/t/informative-priors-to-reduce-prediction-uncertainty/2957 "2020-02-06T03:12:50Z")

</div>

Hello everyone, We have had a hand full of excellent discussion on quantification of uncertainty in prediction models/ risk calculators: here, here. The latter discussion touched briefly on Bayesian approach to predict…

---

## [Bootstrapping CIs in multiply imputed data](https://discourse.datamethods.org/t/bootstrapping-cis-in-multiply-imputed-data/1926)

<div class="topic-metadata">

**Author:** [@S\_Delaney](https://discourse.datamethods.org/u/S_Delaney)\
**Replies:** 5\
**Last updated:** [January 25, 2020, 11:50pm UTC](https://discourse.datamethods.org/t/bootstrapping-cis-in-multiply-imputed-data/1926 "2020-01-25T23:50:30Z")

</div>

Background: I plan to estimate the average causal effect of a continuous exposure on my outcome using the parametric g-formula (i.e., standardized mean difference), and I need to bootstrap the 95% CIs. I’d like to use 10…

---

## [Precision and decision making](https://discourse.datamethods.org/t/precision-and-decision-making/2750)

<div class="topic-metadata">

**Author:** [@paulpharoah](https://discourse.datamethods.org/u/paulpharoah)\
**Replies:** 29\
**Last updated:** [December 12, 2019, 12:42am UTC](https://discourse.datamethods.org/t/precision-and-decision-making/2750 "2019-12-12T00:42:52Z")

</div>

I will fix this post about medical decision making with an example that relates to a manuscript my colleagues and I posted recently on MedRxiv ( https://medrxiv.org/cgi/content/short/19011924v1). Women at high risk of t…

---

## [Robust Estimation of Marginal Effects (and Uncertainty) using Bayesian Additive Regression Trees](https://discourse.datamethods.org/t/robust-estimation-of-marginal-effects-and-uncertainty-using-bayesian-additive-regression-trees/1470)

<div class="topic-metadata">

**Author:** [@dr.vancleve](https://discourse.datamethods.org/u/dr.vancleve)\
**Replies:** 2\
**Last updated:** [March 29, 2019, 6:18pm UTC](https://discourse.datamethods.org/t/robust-estimation-of-marginal-effects-and-uncertainty-using-bayesian-additive-regression-trees/1470 "2019-03-29T18:18:08Z")

</div>

Hi all. Thanks to all for this frequently fascinating board. My research interests at present center on using observational datasets to support causal inference. I’m well aware of the philosophical, analytical, and prac…

---

## [Can Confidence Intervals Indicate Uncertainty/Precision?](https://discourse.datamethods.org/t/can-confidence-intervals-indicate-uncertainty-precision/145)

<div class="topic-metadata">

**Author:** [@zad](https://discourse.datamethods.org/u/zad)\
**Replies:** 2\
**Last updated:** [December 12, 2018, 3:48pm UTC](https://discourse.datamethods.org/t/can-confidence-intervals-indicate-uncertainty-precision/145 "2018-12-12T15:48:52Z")

</div>

A lot of people who interpret confidence intervals (and not as credible intervals) aim to use their span as an indication of precision and uncertainty, however, there have been arguments against this suggesting that conf…

---

## [Confidence score for survival model](https://discourse.datamethods.org/t/confidence-score-for-survival-model/848)

<div class="topic-metadata">

**Author:** [@gensheim](https://discourse.datamethods.org/u/gensheim)\
**Replies:** 2\
**Last updated:** [October 25, 2018, 4:40pm UTC](https://discourse.datamethods.org/t/confidence-score-for-survival-model/848 "2018-10-25T16:40:46Z")

</div>

I am hoping to get advice on how to construct a confidence score for a Cox PH survival model. I used an electronic medical record (EMR) dataset to train a survival model. For each patient, a time-weighted average of all …