Pith. sign in

REVIEW 4 major objections 5 minor 46 references

Hierarchical Temporal Point Process Modeling of Aggressive Behavior Onset in Psychiatric Inpatient Youth with Autism for Branching Factor Estimation

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Aggression in autistic inpatient youth is less contagious than pooled models claim

desk verdict Solid empirical comparison, but 'less biased' is asserted, not demonstrated; the GOF claim is contradicted by their own Table 4. read the letter →

arxiv 2507.12424 v3 pith:LDXNL7QG submitted 2025-07-16 stat.AP math.PRstat.OT

Michael Potter , Michael Everett , Deniz Erdogmus , Yuna Watanabe , Tales Imbiriba , Matthew S. Goodwin This is my paper · ORCID
classification stat.APmath.PRstat.OT
keywords HawkesprocessbranchingfactorhierarchicalmodelaggressivebehaviorautisminpatienttemporalpointpartialpoolingBayesianinference
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that previous estimates of how strongly one aggressive behavior onset triggers the next were inflated, because they pooled all patients into one homogeneous model. The authors fit a hierarchical Hawkes process with an exponential kernel and an edge-effect correction, allowing each patient's own triggering rate to borrow strength from the group. They report a sample-population branching factor of 0.742 ± 0.026, statistically significantly lower than the pooled model's 0.899 ± 0.015, and far more precise than the unpooled model's 0.717 ± 0.139. If correct, this changes the predicted cascade size from about nine descendant onsets per parent onset to about three, shifting clinical emphasis from internal escalation toward external triggers and individualized monitoring.

What carries the argument

The central object is the branching factor α in a Hawkes process with exponential kernel φ(t) = αβ exp(−βt), where α is the expected number of direct offspring per event and the process is subcritical when α < 1. The model adds an edge-effect initial intensity term (μ0 − μ)β exp(−βt) to account for unobserved history before each observation session. The key mechanism is partial pooling: patient-level parameters μn, αn, βn are drawn from LogNormal distributions whose population-level means and scales are learned from the data, so that sparse patients borrow strength from the group while high-frequency patients do not dominate.

What would settle it

Fit the same hierarchical model with a power-law or nonparametric self-excitation kernel and compare the posterior of the population branching factor; if the subcritical estimate near 0.742 moves substantially or crosses into the supercritical regime, the exponential-kernel assumption is the deciding factor and the branching factor estimate is not robust.

Watch

Extended reading notes

Core claim

The central claim is that partially pooling patient-specific Hawkes process parameters corrects a systematic upward bias in the sample-population branching factor for aggressive behavior onset. Using a conditional intensity with baseline plus exponential self-excitation, the hierarchical model yields a mean branching factor of 0.742 with a standard deviation of 0.026, versus 0.899 ± 0.015 for the pooled model and 0.717 ± 0.139 for the unpooled model. The paper further claims this reduces the expected total number of descendants per parent onset from 9.09 ± 1.58 (pooled) to 2.92 ± 0.40 (hierarchical), and that the hierarchical model is robust to prior and likelihood power-scaling perturbations while the unpooled model is not, especially for individuals with sparse data.

Load-bearing premise

The load-bearing premise is that aggressive behavior onsets are generated by a stationary exponential Hawkes process with constant baseline intensity and a single exponential self-excitation kernel; if the true exogenous rate varies with unobserved context or if memory of past events is not exponential, the estimated branching factor and downstream cascade sizes are model artifacts rather than properties of the behavior.

Editorial extensions

If this is right

  • The sample-population branching factor for aggressive behavior onset in this cohort is below the critical value of 1, meaning cascades are not self-sustaining.
  • Each aggressive onset is expected to produce about 2.9 subsequent onsets, roughly three times fewer than the pooled model's 9.1, directly affecting predicted escalation risk.
  • The hierarchical model provides more precise population-level estimates and more stable individual-level estimates than the unpooled model for patients with sparse data.
  • Separating exogenous from endogenous onsets can support linkage of aggressive behavior to physiological signals and individualized early-warning systems.
  • Pooled Hawkes models applied to heterogeneous clinical populations are likely to overstate self-excitation and misclassify exogenous events as endogenous.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The threefold cascade reduction is a direct mathematical consequence of the formula α/(1−α): moving α from about 0.90 to about 0.74 changes expected descendants from 9 to under 3, so the clinical difference is driven entirely by the point estimate of α.
  • A natural extension would be to test the model on data that include known external triggers such as staff changes or medication timing, to validate whether events classified as exogenous actually align with documented environmental events.
  • The choice of exponential kernel was justified by robustness results from financial data, not validated on this population; a power-law or nonparametric kernel comparison would test whether the subcritical estimate is kernel-independent.
  • Individual branching factors from the hierarchical model could be used directly as patient-level risk scores, stratifying youth for different intensities of monitoring or intervention.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proposes a Bayesian hierarchical (partially pooled) Hawkes process with an exponential kernel and edge-effect correction to estimate the population branching factor for aggressive behavior onsets in 70 psychiatric inpatient youth with autism. It compares this model with pooled and unpooled alternatives, reporting a population branching factor of 0.742 ± 0.026 for the hierarchical model, versus 0.899 ± 0.015 for the pooled model and 0.717 ± 0.139 for the unpooled model. From these estimates it derives expected cascade sizes of about 2.92 versus 9.09 descendants per parent onset. The paper also reports MCMC diagnostics, power-scaling sensitivity analyses, PSIS-LOO cross-validation, Lewis tests with Durbin's modification, and RTCT residual analyses, and uses the inferred branching structure to illustrate potential links with physiological signals.

Significance. If the hierarchical estimate is valid, the result is clinically consequential: it would place the population dynamics in a more strongly subcritical regime than previously reported, changing estimated cascade sizes by roughly a factor of three. The paper has genuine strengths: the inference pipeline is careful, with convergence diagnostics, posterior predictive checks, power-scaling sensitivity analysis, and multiple goodness-of-fit criteria; the appendices provide detailed per-parameter diagnostics and individual-level data; and the software choices (NumPyro, ArviZ) support reproducibility. However, the central claim that the hierarchical model is 'less biased' is not validated against any ground truth, and some goodness-of-fit statements are not supported by the reported tables. The contribution is therefore promising but not yet established at the level claimed.

major comments (4)
  1. [Section 3.1, Figure 2] The central claim that partial pooling 'reduces bias from high-frequency individuals' is not supported by any ground-truth comparison. No simulation study, synthetic-data recovery check, or external benchmark is reported; the evidence for bias reduction is that the hierarchical estimate is lower and more precise than the pooled estimate and more stable than the unpooled estimate. A lower estimate is not automatically a less biased estimate: the difference between the pooled and hierarchical posteriors could reflect prior shrinkage under the Gamma(2.5, 0.4) hyperprior on mu_alpha, kernel misspecification, or nonstationarity of the exogenous rate, rather than correction of a known bias. The paper should add a simulation study in which data are generated from a known hierarchical Hawkes process with a known population branching factor and compare the recovery properties of the pooled, unpooled, and partially pooled estimators. This is load-bearing because the entire clinical interpretation rests on the claim that 0.742 is closer to the truth than 0.899.
  2. [Section 3.4, Table 4] The prose states that 'the partially pooled model yields the highest proportion of sessions that do not reject the null hypothesis,' but Table 4 contradicts this. At the session level, the unpooled model has higher non-rejection rates at significance levels 0.05 (0.88 vs. 0.86) and 0.10 (0.83 vs. 0.81); at the person level, the unpooled model is also higher at 0.05 (0.90 vs. 0.86) and 0.10 (0.84 vs. 0.83). The partially pooled model is higher only at the 0.15 level. Similarly, Table 3 shows an ELPD difference of only 4.81 with a standard error of 17.22 between the partially pooled and unpooled models, which is not statistically significant and is much smaller than the reported uncertainty. The claim that goodness-of-fit measures 'consistently favored' the hierarchical model is therefore not supported by the reported evidence. The GOF section should be rewritten to state accurately which comparisons favor which model, and the strength of the evidence should be calibrated accordingly.
  3. [Section 2.2.2, Eqs. (4)-(5)] The model assumes a stationary exponential Hawkes process with a constant baseline intensity and a single exponential excitation kernel. This assumption is load-bearing because the branching factor is a fitted parameter of this assumed model. The paper justifies the exponential kernel by citing Filimonov and Sornette's financial-data robustness results, but no validation is provided for this population, where unobserved context such as staff changes, medication adjustments, or session-specific conditions could make the baseline rate nonstationary, or where memory of past events may not be exponential. If the exogenous rate varies with context, a fitted exponential kernel can absorb that variation into the self-excitation term and inflate alpha; the lower hierarchical estimate could then reflect shrinkage rather than a true property of the behavior. A concrete and feasible check would be a simulation study using a nonstationary baseline or a non-exponential kernel, comparing the estimated branching factor under the three models; alternatively, the authors could fit a model with time-varying or session-level covariates and report whether the branching factor estimate changes materially.
  4. [Section 3.1, Figure 3] The 'threefold smaller cascade' result is a re-expression of the fitted branching factor through the formula mu_alpha/(1 - mu_alpha), not an independent empirical finding. The expected number of descendants inherits any bias or misspecification in the branching factor estimate, so the clinical contrast between 2.92 and 9.09 descendants cannot be presented as additional evidence for the hierarchical model. The paper should either describe this quantity explicitly as a derived summary of the posterior estimate, or provide an out-of-sample or simulation-based validation that the implied cascade sizes are accurate. As written, the cascade comparison is a transformation of the same parameters being compared, and the uncertainty reported for the ratio does not account for model misspecification.
minor comments (5)
  1. [Section 2.5.1] The text contains a typo: 'Postetior Predictive Checks' should be 'Posterior Predictive Checks'.
  2. [Section 3.1] There is a misspelling: 'estaimates' should be 'estimates'.
  3. [Figure 5 caption] The caption lists panels (a) through (f) but the text refers to '(g) Session 41'; the panel labeling should be made consistent.
  4. [Table 4 caption] The caption says that at the person level the table reports 'the average proportion where the null hypothesis was rejected,' but the values (0.86, 0.90, etc.) are consistent with non-rejection rates, not rejection rates; the caption should be corrected.
  5. [Section 2.3] The text states an ESS target of at least 100, while the reported diagnostics show ESS values exceeding 3,000. The target should either be stated as a diagnostic threshold rather than a target for the actual run, or the text should explain why the run exceeded it by such a large margin.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular derivation chain: branching-factor estimates are fitted targets and the cascade-size claim is an explicitly stated mathematical transform; self-citations are background, not load-bearing.

full rationale

The paper's central quantity, the sample population branching factor, is a model parameter estimated from the observed onset times via MCMC, not a quantity derived from the claim it is used to support. The pooled versus hierarchical comparison is an empirical model comparison on the data (Section 3.1). The 'threefold smaller' expected cascade size is computed from the fitted branching factor using the standard identity μ/(1−μ), and the paper explicitly says this 'translates to' the estimate, so no predicted quantity is passed off as independent. Sensitivity power-scaling perturbs priors and likelihoods within the assumed model; it does not validate the model, which is a limitation but not circularity. The paper cites its own prior work [16] for the pooled baseline, data preprocessing, and the observation that pooled models ignore heterogeneity; these citations are background and not load-bearing for the hierarchical derivation. No uniqueness theorem, kernel identity, or central premise is imported from the authors' prior work. The exponential-kernel choice is justified by an external citation [21], and the absence of kernel validation on this population is a correctness risk, not circular self-support. Internal inconsistency: Table 4 does not support the prose claim that the hierarchical model has the highest non-rejection rates at 0.05 and 0.10 (the unpooled model is higher), and Sections 3.3 and 4.2 acknowledge that the physiological-linkage and generalizability findings are preliminary and limited. These issues affect evidentiary strength, not circularity.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The central estimate rests on the exponential Hawkes specification, the independence of sessions, and the chosen weakly informative priors. No new physical or causal entity is introduced; the latent branching structure is a model construct, not an independently evidenced entity. The main free choices are the kernel form, the prior scales, and the per-session edge-effect correction.

free parameters (4)
  • Population branching factor hyperprior mu_alpha ~ Gamma(2.5, 0.4) = shape 2.5, scale 0.4 (prior mean 1.0, median ~0.87)
    Chosen by authors, not fit to data; sets the center of the hierarchical distribution and influences sparse individuals, although the posterior moves to 0.742.
  • Population-level hyperpriors for baseline and decay, and scale parameters = Half-Normal(0.1), Half-Cauchy(1.5), Half-Cauchy(1.0), Half-Cauchy(1.5)
    Weakly informative choices recommended by [26]; power-scaling checks show relative robustness for the branching factor, but likelihood sensitivity values up to 0.51 are not negligible.
  • Session-level edge-effect increments Delta_mu_in = Posterior means not reported for all sessions; mu_0 per session
    Each session adds an initial intensity correction; necessary to avoid misclassifying first events as exogenous, but it is an extra fitted quantity per session.
  • Person-level parameters mu_n, alpha_n, beta_n for 70 participants = Posterior means reported in Appendix D tables
    These are the fitted individual branching factors, baselines, and decay rates; the hierarchical prior regularizes them through population-level hyperpriors.
assumptions (5)
  • domain assumption Aggressive onsets follow a Hawkes process with conditional intensity lambda(t) = mu + sum alpha beta exp(-beta(t-t_j)) plus edge-effect term (mu_0 - mu) beta exp(-beta t)
    Introduced in Section 2.2.2, Eqs. (4) and (5); all branching factor estimates and cascade sizes inherit this assumption. The exponential kernel is justified by a citation from financial data rather than validated on this population.
  • domain assumption Observation sessions are independent given person parameters, and the joint likelihood factorizes over 429 sessions
    Stated in Section 2.2.1; if carry-over agitation or other dependencies link sessions, the likelihood is misspecified.
  • domain assumption Weakly informative priors are adequate and do not force the branching factor estimate
    Chosen in Section 2.2.3 and Appendix B; power-scaling checks suggest robustness for the hierarchical mean, but a sensitivity check is not a proof of prior adequacy.
  • standard math Random Time Change Theorem and plug-in estimated intensity can be used for goodness-of-fit testing
    Used in Section 2.5; the RTCT holds for the true intensity, but using a point estimate of theta can distort nominal test sizes, which the paper partially acknowledges in Appendix C.
  • ad hoc to paper Welch's t-test on posterior draws is a valid way to compare branching factor estimates
    Section 3.1 uses Welch's t-test on 4000 MCMC samples; posterior draws are correlated and the test assumes independent samples, so p-values should be treated as approximate.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Hierarchical Temporal Point Process Modeling of Aggressive Behavior Onset in Psychiatric Inpatient Youth with Autism for Branching Factor Estimation." pith.science (2026). https://pith.science/paper/LDXNL7QG

@misc{pith2026250712424,
  author       = {Pith},
  title        = {Pith review of: Hierarchical Temporal Point Process Modeling of Aggressive Behavior Onset in Psychiatric Inpatient Youth with Autism for Branching Factor Estimation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LDXNL7QG}},
  note         = {Machine review of arXiv:2507.12424}
}
read the original abstract

Aggressive behavior in autistic inpatient youth often arises in temporally clustered bursts complicating efforts to distinguish external triggers from internal escalation. The sample population branching factor-the expected number of new onsets triggered by a given event-is a key summary of self-excitation in behavior dynamics. Prior pooled models overestimate this quantity by ignoring patient-specific variability. We addressed this using a hierarchical Hawkes process with an exponential kernel and edge-effect correction allowing partial pooling across patients. This approach reduces bias from high-frequency individuals and stabilizes estimates for those with sparse data. Bayesian inference was performed using the No U-Turn Sampler with model evaluation via convergence diagnostics, power-scaling sensitivity analysis, and multiple Goodness-of-Fit (GOF) metrics: PSIS-LOO the Lewis test with Durbin's modification and residual analysis based on the Random Time Change Theorem (RTCT). The hierarchical model yielded a significantly lower and more precise branching factor estimate mean (0.742 +- 0.026) than the pooled model (0.899 +- 0.015) and narrower intervals than the unpooled model (0.717 +- 0.139). This led to a threefold smaller cascade of events per onset under the hierarchical model. Sensitivity analyses confirmed robustness to prior and likelihood perturbations while the unpooled model showed instability for sparse individuals. GOF measures consistently favored or on par to the hierarchical model. Hierarchical Hawkes modeling with edge-effect correction provides robust estimation of branching dynamics by capturing both within- and between-patient variability. This enables clearer separation of endogenous from exogenous events supports linkage to physiological signals and enhances early warning systems individualized treatment and resource allocation in inpatient care.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

46 extracted references · 46 canonical work pages

  1. [1]

    Prevalence and characteristics of autism spectrum disorder among children aged 8 years—autism and developmental disabilities monitoring network, 11 sites, United States, 2020

    Maenner MJ. Prevalence and characteristics of autism spectrum disorder among children aged 8 years—autism and developmental disabilities monitoring network, 11 sites, United States, 2020. MMWR Surveillance Summaries. 2023;72

    work page 2020

  2. [2]

    The occurrence of challenging behaviours in chil- dren with autism spectrum disorders and atypical development

    Hattier MA, Matson JL, Belva BC, Horovitz M. The occurrence of challenging behaviours in chil- dren with autism spectrum disorders and atypical development. Developmental Neurorehabilitation. 2011;14(4):221–229

    work page 2011

  3. [3]

    Aggression in children and adolescents with ASD: Prevalence and risk factors

    Kanne SM, Mazurek MO. Aggression in children and adolescents with ASD: Prevalence and risk factors. Journal of autism and developmental disorders. 2011;41:926–937

    work page 2011

  4. [4]

    Assessing aggression in persons with autism spectrum disorders: An overview

    Matson JL, Cervantes PE. Assessing aggression in persons with autism spectrum disorders: An overview. Research in developmental disabilities. 2014;35(12):3269–3275. 17

    work page 2014

  5. [5]

    Parent-defined target symptoms respond to risperidone in RUPP autism study: customer approach to clinical trials

    Arnold LE, Vitiello B, McDougle C, Scahill L, Shah B, Gonzalez NM, et al. Parent-defined target symptoms respond to risperidone in RUPP autism study: customer approach to clinical trials. Journal of the American Academy of Child & Adolescent Psychiatry. 2003;42(12):1443–1450

    work page 2003

  6. [6]

    A comparison of health care utilization and costs of children with and without autism spectrum disorders in a large group-model health plan

    Croen LA, Najjar DV , Ray GT, Lotspeich L, Bernal P. A comparison of health care utilization and costs of children with and without autism spectrum disorders in a large group-model health plan. Pediatrics. 2006;118(4):e1203–e1211

    work page 2006

  7. [7]

    Emotion regulation: Concepts & practice in autism spectrum disorder

    Mazefsky CA, White SW. Emotion regulation: Concepts & practice in autism spectrum disorder. Child and adolescent psychiatric clinics of North America. 2013;23(1):10–1016

    work page 2013

  8. [8]

    Conducting research with minimally verbal participants with autism spectrum disorder

    Tager-Flusberg H, Plesa Skwerer D, Joseph RM, Brukilacchio B, Decker J, Eggleston B, et al. Conducting research with minimally verbal participants with autism spectrum disorder. Autism. 2017;21(7):852–861

    work page 2017

Show all 46 references
  1. [9]

    Parenting stress in mothers and fathers of toddlers with autism spectrum disorders: Associations with child characteristics

    Davis NO, Carter AS. Parenting stress in mothers and fathers of toddlers with autism spectrum disorders: Associations with child characteristics. Journal of autism and developmental disorders. 2008;38:1278– 1291

    2008

  2. [10]

    Home sweet home? Families’ experiences with aggression in children with autism spectrum disorders

    Hodgetts S, Nicholas D, Zwaigenbaum L. Home sweet home? Families’ experiences with aggression in children with autism spectrum disorders. Focus on autism and other developmental disabilities. 2013;28(3):166–174

    2013

  3. [11]

    Violence faced by staff in a learning disability service

    Kiely J, Pankhurst H. Violence faced by staff in a learning disability service. Disability and Rehabilitation. 1998;20(3):81–89

    1998

  4. [12]

    Wearable biosensing to predict imminent aggressive behavior in psychiatric inpatient youths with autism

    Imbiriba T, Demirkaya A, Singh A, Erdogmus D, Goodwin MS. Wearable biosensing to predict imminent aggressive behavior in psychiatric inpatient youths with autism. JAMA network open. 2023;6(12):e2348898–e2348898

    2023

  5. [13]

    Behavioral momentum theory: Equations and applications

    Nevin JA, Shahan TA. Behavioral momentum theory: Equations and applications. Journal of Applied Behavior Analysis. 2011;44(4):877–895

    2011

  6. [14]

    Treatment relapse and behavioral momentum theory

    Pritchard D, Hoerger M, Mace FC. Treatment relapse and behavioral momentum theory. Journal of applied behavior analysis. 2014;47(4):814–833

    2014

  7. [15]

    Assessing challenging behaviors in Autism Spectrum Disorders: Prevalence, rating scales, and autonomic indicators

    Cohen IL, Yoo JH, Goodwin MS, Moskowitz L. Assessing challenging behaviors in Autism Spectrum Disorders: Prevalence, rating scales, and autonomic indicators. International handbook of autism and pervasive developmental disorders. 2011;p. 247–270

    2011

  8. [16]

    Temporal Point Process Modeling of Aggressive Behavior Onset in Psychiatric Inpatient Youths with Autism

    Potter M, Everett M, Singh A, Stratis G, Watanabe Y , Demirkaya A, et al. Temporal Point Process Modeling of Aggressive Behavior Onset in Psychiatric Inpatient Youths with Autism. arXiv preprint arXiv:250315821. 2025

    2025

  9. [17]

    Hawkes models and their applications

    Laub PJ, Lee Y , Pollett PK, Taimre T. Hawkes models and their applications. Annual Review of Statistics and Its Application. 2024;12

    2024

  10. [18]

    Exogenous and endogenous factors affecting stock market transactions: A Hawkes process analysis of the Tokyo Stock Exchange during the COVID-19 pandemic

    Ito MI, Honma Y , Ohnishi T, Watanabe T, Aihara K. Exogenous and endogenous factors affecting stock market transactions: A Hawkes process analysis of the Tokyo Stock Exchange during the COVID-19 pandemic. Plos one. 2024;19(4):e0301462

    2024

  11. [19]

    A systematic review of physiological reactivity to stimuli in autism

    Lydon S, Healy O, Reed P, Mulhern T, Hughes BM, Goodwin MS. A systematic review of physiological reactivity to stimuli in autism. Developmental neurorehabilitation. 2016;19(6):335–355

    2016

  12. [20]

    Lecture notes: Temporal point processes and the conditional intensity function

    Rasmussen JG. Lecture notes: Temporal point processes and the conditional intensity function. arXiv preprint arXiv:180600221. 2018

    2018

  13. [21]

    Apparent criticality and calibration issues in the Hawkes self-excited point process model: application to high-frequency financial data

    Filimonov V , Sornette D. Apparent criticality and calibration issues in the Hawkes self-excited point process model: application to high-frequency financial data. Quantitative Finance. 2015;15(8):1293– 1314. 18

    2015

  14. [22]

    The elements of Hawkes processes

    Laub PJ, Lee Y , Taimre T. The elements of Hawkes processes. Springer; 2021

    2021

  15. [23]

    Context-aware experience sampling reveals the scale of variation in affective experience

    Hoemann K, Khan Z, Feldman MJ, Nielson C, Devlin M, Dy J, et al. Context-aware experience sampling reveals the scale of variation in affective experience. Scientific reports. 2020;10(1):12459

    2020

  16. [24]

    Multilevel (hierarchical) modeling: what it can and cannot do

    Gelman A. Multilevel (hierarchical) modeling: what it can and cannot do. Technometrics. 2006;48(3):432–435

    2006

  17. [25]

    Comparing Bayesian models of annotation

    Paun S, Carpenter B, Chamberlain J, Hovy D, Kruschwitz U, Poesio M. Comparing Bayesian models of annotation. Transactions of the Association for Computational Linguistics. 2018;6:571–585

    2018

  18. [26]

    Prior distributions for variance parameters in hierarchical models (comment on article by Browne and Draper)

    Gelman A. Prior distributions for variance parameters in hierarchical models (comment on article by Browne and Draper). Bayesian Anal. 2006

    2006

  19. [27]

    The No-U-Turn sampler: adaptively setting path lengths in Hamiltonian Monte Carlo

    Hoffman MD, Gelman A, et al. The No-U-Turn sampler: adaptively setting path lengths in Hamiltonian Monte Carlo. J Mach Learn Res. 2014;15(1):1593–1623

    2014

  20. [28]

    Bayesian data analysis

    Gelman A, Carlin JB, Stern HS, Rubin DB. Bayesian data analysis. Chapman and Hall/CRC; 1995

    1995

  21. [29]

    Rank-normalization, folding, and local- ization: An improved R-hat for assessing convergence of MCMC (with discussion)

    Vehtari A, Gelman A, Simpson D, Carpenter B, Burkner PC. Rank-normalization, folding, and local- ization: An improved R-hat for assessing convergence of MCMC (with discussion). Bayesian analysis. 2021;16(2):667–718

    2021

  22. [30]

    Efficient non-parametric Bayesian Hawkes processes

    Zhang R, Walder C, Rizoiu MA, Xie L. Efficient non-parametric Bayesian Hawkes processes. arXiv preprint arXiv:181003730. 2018

    2018

  23. [31]

    The power of alternative Kolmogorov-Smirnov tests based on transformations of the data

    Kim SH, Whitt W. The power of alternative Kolmogorov-Smirnov tests based on transformations of the data. ACM Transactions on Modeling and Computer Simulation (TOMACS). 2015;25(4):1–22

    2015

  24. [32]

    Practical Bayesian model evaluation using leave-one-out cross-validation and W AIC

    Vehtari A, Gelman A, Gabry J. Practical Bayesian model evaluation using leave-one-out cross-validation and W AIC. Statistics and computing. 2017;27:1413–1432

    2017

  25. [33]

    Performances of LOO and W AIC as IRT model selection methods

    Luo Y , Al-Harbi K. Performances of LOO and W AIC as IRT model selection methods. Psychological Test and Assessment Modeling. 2017;59(2):183

    2017

  26. [34]

    Implicitly adaptive importance sampling

    Paananen T, Piironen J, B ¨urkner PC, Vehtari A. Implicitly adaptive importance sampling. Statistics and Computing. 2021;31(2):16

    2021

  27. [35]

    The importance of prior sensitivity analysis in Bayesian statistics: demonstrations using an interactive Shiny App

    Depaoli S, Winter SD, Visser M. The importance of prior sensitivity analysis in Bayesian statistics: demonstrations using an interactive Shiny App. Frontiers in psychology. 2020;11:608045

    2020

  28. [36]

    Detecting and diagnosing prior and likelihood sensitivity with power-scaling

    Kallioinen N, Paananen T, B ¨urkner PC, Vehtari A. Detecting and diagnosing prior and likelihood sensitivity with power-scaling. Statistics and Computing. 2024;34(1):57

    2024

  29. [37]

    Critical reflexivity in financial markets: a Hawkes process analysis

    Hardiman SJ, Bercot N, Bouchaud JP. Critical reflexivity in financial markets: a Hawkes process analysis. The European Physical Journal B. 2013;86:1–9

    2013

  30. [38]

    Branching processes: their role in epidemiology

    Jacob C. Branching processes: their role in epidemiology. International journal of environmental research and public health. 2010;7(3):1186–1204

    2010

  31. [39]

    Statistical models for earthquake occurrences and residual analysis for point processes

    Ogata Y . Statistical models for earthquake occurrences and residual analysis for point processes. Journal of the American Statistical association. 1988;83(401):9–27

    1988

  32. [40]

    First-and second-order statistics characterization of Hawkes processes and non- parametric estimation

    Bacry E, Muzy JF. First-and second-order statistics characterization of Hawkes processes and non- parametric estimation. IEEE Transactions on Information Theory. 2016;62(4):2184–2202

    2016

  33. [41]

    Accessed: 2025-07-

    DrivenData.: DrivenData: Data Science & AI Competitions to build a better world. Accessed: 2025-07-

    2025

  34. [42]

    https://www.drivendata.org/

  35. [43]

    Composable effects for flexible and accelerated probabilistic programming in NumPyro

    Phan D, Pradhan N, Jankowiak M. Composable effects for flexible and accelerated probabilistic programming in NumPyro. arXiv preprint arXiv:191211554. 2019;. 19

    2019

  36. [44]

    Available from: http://github.com/jax-ml/jax

    Bradbury J, Frostig R, Hawkins P, Johnson MJ, Leary C, Maclaurin D, et al.: JAX: composable transformations of Python+NumPy programs. Available from: http://github.com/jax-ml/jax

  37. [45]

    ArviZ a unified library for exploratory analysis of Bayesian models in Python

    Kumar R, Carroll C, Hartikainen A, Martin O. ArviZ a unified library for exploratory analysis of Bayesian models in Python. Journal of Open Source Software. 2019;4(33):1143. https://doi.org/10. 21105/joss.01143

    2019

  38. [46]

    prior” and “likelihood

    Mason DM, Schuenemeyer JH. A modified Kolmogorov-Smirnov test sensitive to tail alternatives. The annals of Statistics. 1983;p. 933–946. 20 Appendix A Individual Data Statistics Individual ID # Of Onsets Total Observation Time (min) # of Sessions Time (min) / Session 0 1287 20...

    1983

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.