<b>MCMCpack</b>: Markov Chain Monte Carlo in<i>R</i>

Martin, Andrew D.; Quinn, Kevin M.; Park, Jong Hee

doi:10.18637/jss.v042.i09

Cited by 540 publications

(411 citation statements)

References 49 publications

Supporting

Mentioning

402

Contrasting

Unclassified

Order By: Relevance

“…Bayesian analyses were computed using WinBUGS (Lunn et al, 2000), and the packages MCMCpack (Martin et al, 2011) and R2WinBUGS (Sturtz et al, 2005) in R 3.0.1 (R Core Team, 2013). The Bayesian approach allows obtaining a posterior distribution of the parameters (e.g., regression slopes) given the data and existing knowledge (priors) for these parameters (McCarthy and Masters, 2005;McCarthy, 2007).…”

Section: Discussionmentioning

confidence: 99%

“…This represents a clear advantage, because allows to directly compute the probability that a parameter has a particular value rather than the probability to obtain a particular type of data given some null hypothesis (Ke´ry, 2010). The posterior distributions were obtained with the computation of Markov chain Monte Carlo (MCMC), which is the method commonly used to fit models using a Bayesian approach (Martin et al, 2011). For each analysis, a Markov chain of length 11 000 was estimated, discarding the first 1000 points as a burn-in.…”

Section: Discussionmentioning

confidence: 99%

“…Figs. 3 and 4) was tested with the Bayes factors tool in the MCMCpack (Martin et al, 2011). In each run, three models were tested, including a unique intercept and slope, different intercepts and a common slope, and different intercepts and slopes.…”

Section: Discussionmentioning

confidence: 99%

“…In each lake, the models were almost coincident considering separately the two layers (0 m and 20 m). In Lake Garda, the comparison by Bayes Factor (Martin et al, 2011) of the regression model with a common slope and intercept, versus models with different intercepts and intercepts/slopes showed that there was barely positive and positive evidence to support the simpler model (Bayes factors, 1.02 and 3.9, respectively) (Fig. 3a).…”

Section: Mcs -Planktothrix Biovolume Modelsmentioning

confidence: 99%

See 3 more Smart Citations

Strict coupling between the development ofPlanktothrix rubescensand microcystin content in two nearby lakes south of the Alps (lakes Garda and Ledro)

Salmaso

Adriano

Shams

et al. 2013

Ann. Limnol. - Int. J. Lim.

View full text Add to dashboard Cite

-Cyanobacteria and their principal cyanotoxins were studied in the largest Italian lake (Lake Garda, 65 m a.s.l.) and in a near smaller reservoir (Lake Ledro, 652 m a.s.l.). The two lakes share a fraction of water due to a pipe pumping water from and into the lakes and the same dominant cyanobacterial species (Planktothrix rubescens). Despite the differences in the concentrations of cyanotoxins (mostly microcystins, MCs) and biovolumes of P. rubescens (BV Pr ) (over one order of magnitude), the Bayesian analyses linking these two variables showed striking similarities, suggesting the existence of similar toxic genotypes colonizing the two water bodies and a constitutive MC production. It was stressed that a greater sensitivity and reliability in the management strategies aimed at minimizing the risks due to cyanobacteria should also contemplate the use of specific lake-tailored models linking MCs and BV Pr .

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Mcs -Planktothrix Biovolume Modelsmentioning

confidence: 99%

See 2 more Smart Citations

Strict coupling between the development ofPlanktothrix rubescensand microcystin content in two nearby lakes south of the Alps (lakes Garda and Ledro)

Salmaso

Adriano

Shams

et al. 2013

Ann. Limnol. - Int. J. Lim.

View full text Add to dashboard Cite

show abstract

“…For categorical data with more than two categories (e.g., classifying participants to strategies A, B, and C), we modeled the probabilities i of belonging to the ith category based on the observed category counts using the multinomial functionassuming a uniform Dirichlet prior on the i s; those posteriors were directly approximated with the R function "MCmultinomdirichlet" from the R package "MCMCpack" (Martin, Quinn, & Park, 2011) using 1,000,000 samples.…”

Section: Appendix Amentioning

confidence: 99%

Think twice and then: Combining or choosing in dialectical bootstrapping?

Herzog¹,

Hertwig²

2014

Journal of Experimental Psychology: Learning, Memory, and Cogni

View full text Add to dashboard Cite

Individuals can partly recreate the "wisdom of crowds" within their own minds by combining nonredundant estimates they themselves have generated. Herzog and Hertwig (2009) showed that this accuracy gain could be boosted by urging people to actively think differently when generating a 2nd estimate ("dialectical bootstrapping"). Although the "crowd within" promises accuracy gains, it remains unclear whether and when people spontaneously reap those gains. What makes people combine their estimates rather than trying to identify the better one? This research found that participants were more likely to combine when they were instructed to actively contradict themselves. Furthermore, they were more likely to combine as the size of the disagreement between 1st and 2nd estimate grew. People thus acted as if they were hedging against the risk of making large errors. Finally, when people pursued a strategy other than combination, they were not able to outperform their crowd within.

show abstract

Matched vs Nonmatched Placebos in a Randomized Trial of COVID-19 Treatments

Reis,

Savassi,

Ferreira

et al. 2024

JAMA Netw Open

View full text Add to dashboard Cite

ImportanceMatched placebo interventions are complex and resource intensive. Recent evidence suggests matched placebos may not always be necessary. Previous studies have predominantly evaluated potential bias of nonmatched placebos (ie, differing on dose, frequency of administration, or formulation) in pain and mental health, but to date no systematic examination has been conducted in infectious disease.ObjectiveTo test for differences between nonmatched and matched placebo arms with respect to clinical outcome measures across multiple therapeutics for COVID-19.Design, Setting, and ParticipantsIn a comparative effectiveness research study, a post hoc analysis was conducted of data on individual patients enrolled in a large, multiarm, platform randomized clinical trial in symptomatic adult outpatients with COVID-19 between January 15, 2021, to September 28, 2023, in which the outcomes of both matched and nonmatched placebo groups were reported. Bayesian and frequentist covariate-adjusted techniques were compared with 7 intervention-placebo pairs.ExposuresSeven matched and nonmatched placebo pairs (for a total of 7 comparisons) were evaluated throughout the primary platform trial. Comparisons were made between treatment and its associated matched (concurrent) placebo, as well as with nonmatched placebo (alone and in combination) assessed at a similar time point.Main Outcomes and MeasuresOutcomes assessed included hospitalizations, EuroQol 5-Dimension 5-level scores, and PROMIS Global-10 scores.ResultsA total of 7 intervention-control pairs (N = 2684) were assessed, including 1620 (60.4%) women, with mean (SD) age, 47 (15.2) years; the most common comorbidities were obesity (41.9%) and hypertension (37.9%). In a meta-analysis with decoupled SEs, accounting for overlapping placebo patients, the overall odds ratio (OR) of nonmatched compared with matched placebo was 1.01 (95% credible interval, 0.77-1.32), with posterior probability of equivalence, defined as 0.8 ≤ OR ≤ 1.2 (a deviation from perfect equivalence ie, OR = 1, by no more than 0.2) of 85.4%, implying no significant difference. Unadjusted analysis of the event rate difference between all nonmatched and matched placebo groups did not identify any notable differences across all 7 treatment-placebo combinations assessed. Similar analysis that was conducted for patient-reported quality of life outcomes did not yield statistically significant differences.Conclusions and RelevanceIn this post hoc study of a randomized clinical platform trial, pooling matched and nonmatched placebo patient data did not lead to inconsistencies in treatment effect estimation for any of the investigational drugs. These findings may have significant implications for future platform trials, as the use of nonmatched placebo may improve statistical power, or reduce barriers to placebo implementation.

show abstract

MCMCpack: Markov Chain Monte Carlo inR

Cited by 540 publications

References 49 publications

Strict coupling between the development ofPlanktothrix rubescensand microcystin content in two nearby lakes south of the Alps (lakes Garda and Ledro)

Strict coupling between the development ofPlanktothrix rubescensand microcystin content in two nearby lakes south of the Alps (lakes Garda and Ledro)

Think twice and then: Combining or choosing in dialectical bootstrapping?

Matched vs Nonmatched Placebos in a Randomized Trial of COVID-19 Treatments

Contact Info

Product

Resources

About