Pith. sign in

REVIEW 4 major objections 6 minor 1 cited by

Modeling Human Spatial Mobility Patterns with the L\'evy Flight Cluster Model

T0 review · 4 major / 6 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read The paper introduces the Lévy Flight Cluster Model, a generative Bayesian mixture that turns irregular GPS records into per-person activity regions, long jumps, and returns, with the posterior serving as a simulator of realistic synthetic t

desk verdict The LFCM is a real methodological contribution for activity-space estimation from irregular GPS—just don't over-read the synthetic-trajectory claims until the temporal assumptions get tested. read the letter →

arxiv 2509.00298 v1 pith:YANC2OCG submitted 2025-08-30 stat.ME stat.AP

Malcolm Wolff , Adrian Dobra , Anton H. Westveld , Grace S. Chiu This is my paper · ORCID
classification stat.MEstat.AP MSC 62F1562M0560G51
keywords LévyflighthumanmobilityactivityspaceBayesianmixturemodelGPSdatagenerativeanonymizationoverlap
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

The paper introduces the Lévy Flight Cluster Model (LFCM), a generative Bayesian mixture that takes patchy, irregularly timed GPS records from a person's devices and separates their movement into local wandering within stable activity regions (home, work, a gym), occasional long jumps between regions, and deliberate returns to previously visited regions. The model is designed so that its posterior distribution acts as a generative representation of the person's activity space, letting researchers simulate plausible synthetic trajectories, attach uncertainty bounds to activity regions, and compare different people's mobility even when their devices were recorded at different times. On a simulated daily routine and on 12 weeks of real device data from 293 individuals, the paper reports that samples from the model reproduce the standard human-mobility metrics — jump length, mean-squared displacement, radius of gyration, and frequented locations — and that activity regions stabilize after a few weeks of data. That matters because the same machinery offers a route to data anonymization and to probabilistic overlap (contact) matrices that do not depend on arbitrary spatial grids.

What carries the argument

The LFCM mixture density (Eq. 23) is the load-bearing object: Brownian components with per-activity drift µg and covariance Σ*g scaled by elapsed time Δt, plus a jump component with Pareto lengths, von Mises angles, and a return term that places normal density at the posterior mean and covariance of each previously identified contiguous Brownian path. Four latent indicator vectors carry the classification — b (jump vs. Brownian), c (activity group), η (return vs. exploration), z (activity region) — and a collapsed MCMC sampler integrates out the Normal-Wishart, Dirichlet, Beta, and Pareto-Gamma parameters so the chain explores the discrete allocation space, with absorption/ejection proposals

What would settle it

Fit the LFCM to high-frequency GPS records with known activity labels (e.g., travel diaries): if the posterior allocation of time to activity regions versus jumps systematically misestimates observed dwell times, or if re-fitting the same data subsampled to 5, 15, and 60-minute intervals shifts the estimated activity share substantially, the instantaneous-jump assumption is doing the work. A direct check is regressing observed jump length on elapsed time between consecutive records — the flight assumption predicts no relationship once the mixture allocation is accounted for, whereas a Lévy wal

Watch

Extended reading notes

Core claim

Definition 5.1 (Eq. 23) gives the model: a trajectory's first differences mix Brownian motion within one of NG activity groups (probability 1−ν) with jumps (probability ν) whose lengths are Pareto and directions von Mises; each jump is exploratory or, with probability p, a return to a previously visited activity region. Conjugate priors and a collapsed MCMC sampler integrate out nuisance parameters, leaving a chain over latent allocations b, c, η, z and the group count NG. The paper's central object is the resulting joint posterior, used generatively: it reproduces jump-length, MSD, and radius-of-gyration statistics on 293 devices' GPS records, gives exploration growth ~t^0.52 versus the lit

Load-bearing premise

The model assumes each long move is an instantaneous Lévy jump whose probability does not depend on elapsed time, even though real travel takes time — so the model's split of a person's time between activity and travel rests on a premise the data do not verify.

Editorial extensions

If this is right

  • Irregular, application-driven GPS sampling need not be binned onto grids: the LFCM estimates per-person probabilistic activity regions with uncertainty bounds directly from observed times and locations.
  • Posterior samples generate synthetic trajectories that preserve jump-length, MSD, radius-of-gyration, and frequented-location statistics, offering a practical anonymization route for sharing mobility data.
  • Because trajectories can be interpolated and extrapolated to a common time window from posterior samples, pairwise distances between individuals can be computed even when their device data were recorded at different times, yielding probabilistic contact matrices.
  • Activity-region estimates stabilize quickly — roughly 50% overlap of top regions with 12-week estimates from just 2 weeks of data, rising to about 60% at 4 weeks — so short observation windows can support downstream inference.

Reading between the lines

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

  • None of the model's components is human-specific; the same mixture of Brownian clusters, Lévy jumps, and returns should transfer to animal telemetry or transit data, where per-individual uncertainty around activity regions is similarly needed.
  • If the instantaneous-jump (Lévy flight) assumption is wrong for travel that occupies real time, then subsampling high-frequency GPS to coarser intervals should change the estimated activity-versus-travel time split; if it does, a Lévy-walk variant with speed-dependent jump durations would be needed.
  • The model's sensitivity to the minimal-jump-distance ε suggests a natural upgrade — a prior or empirical-Bayes estimate of ε from the data — which would make the method fully unsupervised and remove the calibration step.
  • The probabilistic overlap matrices are presently pairwise-distance summaries; coupling them with network models could turn them into inference tools for social structure, segregation, or disease-contact risk.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper introduces the Lévy Flight Cluster Model (LFCM), a hierarchical Bayesian mixture model for individual human mobility from temporally irregular GPS observations. The model represents spatial increments as a mixture of Brownian-motion activity clusters and Lévy jumps, augmented by a return mechanism to previously identified activity regions. Posterior inference is performed with a collapsed MCMC sampler, and the estimated posterior is used to generate synthetic trajectories, estimate activity regions, and construct probabilistic overlap matrices. The paper reports a simulation study and an application to 293 mobile devices observed over 12 weeks, comparing mobility metrics (jump length, MSD, radius of gyration) with a grid-based conservative proportional time (CPT) estimator.

Significance. If the model's generative assumptions were valid, the LFCM would be a useful contribution to activity-space estimation, anonymized data generation, and probabilistic contact networks. The collapsed MCMC sampler with conjugate priors is computationally attractive, and the comparison with a grid-based CPT estimator is a sensible baseline. The model's ability to produce uncertainty-aware activity regions from sparse, irregular GPS is potentially valuable. However, the temporal misspecification of the jump component and an inconsistency in the return-region distribution affect the central claims about generative realism and the headline applications. These issues are substantial but appear addressable within the manuscript's scope.

major comments (4)
  1. [Section 5, Eqs. (15)–(17), (23); Table 1] The jump component is temporally misspecified. In a Lévy process, the jump intensity must scale with interval length: for a compound Poisson process, P(N(Δt)>0)=1−exp(−λΔt) and the distribution of the aggregate jump depends on Δt. Eq. (17) writes f_J*(Δr; α, Δt), but the implemented prior and likelihood put a Pareto distribution on ||Δx||_2 alone, with no dependence on Δt, and ν is treated as a constant per observation. Because the GPS data are explicitly temporally irregular (Section 2), this cannot correctly allocate time between activity and travel. The Brownian variance scaling does not repair the issue: it can absorb long gaps as large Brownian displacements, but it does not capture the increased opportunity for long trips over long intervals. This directly affects the synthetic-trajectory and overlap-matrix applications, where travel time is misrepresented as instantaneous teleport
  2. [Section 5.1, Eqs. (22)–(23); Appendix A] The return-region distribution is inconsistent. Eq. (22) derives the distribution of the time-average of a Brownian motion over [0,T], with mean μ0+T/2·diag(μ) and variance (T/3)·diag(σ²). This is an average location, not an endpoint distribution. Yet Eq. (23) uses N(Δx(t); rμ_z, Σ_z) inside a model for increments, while Appendix A uses N(x(t_i); rμ_z, Σ_z) for the absolute location. A displacement and an absolute activity-region center are not interchangeable, and the time-average distribution is not the appropriate likelihood for a return point. This affects the classification of returns and the geometry of estimated activity regions. The authors should either model the absolute location with a properly derived endpoint distribution, or explicitly define the activity-region center as a time-average and derive the resulting return likelihood.
  3. [Section 7, Table 2] The model validation is in-sample and partly circular. The metrics compared in Table 2 — jump length, MSD, and radius of gyration — are the same heavy-tailed and clustering features that the LFCM was constructed to reproduce. The table reports only means and standard deviations, without uncertainty intervals or a comparison of full distributions. The pooled tail index α≈1.71–1.74 estimated in Section 3 is obtained from the same dataset and is not used as an independent prediction. The claim that the LFCM 'accurately captures the key characteristics of human mobility' therefore needs stronger support, for example by fitting the model on early weeks and simulating later weeks, or by reporting posterior predictive checks against held-out data.
  4. [Section 6, 'Data Generation'] The simulation study does not test the temporal allocation that is central to the model's intended applications. The simulated data include finite-speed travel — e.g., a 20-minute commute between home and work — while the LFCM assumes instantaneous Lévy jumps. The simulation demonstrates recovery of activity regions, but it does not evaluate whether the model correctly apportions time between activity regions and travel. At a minimum, the authors should report time spent in travel versus time spent in activity regions for LFCM-generated trajectories compared with the true simulation, and ideally include a simulation scenario that matches the Lévy-flight assumption to separate the model's spatial clustering performance from its temporal realism.
minor comments (6)
  1. [Throughout] There are numerous typographical errors: 'L´evy' spacing inconsistencies, 'Mendelbrot' (Section 4), 'noticably' (Section 7), and 'activity activity spaces' (Section 7).
  2. [Eq. (24)] The definitions of A and B in Eq. (24) are hard to parse due to nested indicators and line breaks. Rewriting with explicit indicator products would improve readability.
  3. [Table 1] The prior for N_G is listed as Poiss(λ) with Gamma(1/2,1/2) hyperprior, but the text in Section 5.2 says the prior N_G ~ Gamma(1/2,1/2). Also, Table 1 uses α_g for the Pareto parameter while Eq. (23) uses a common α; clarify which is intended.
  4. [Section 5.2 and Algorithm 2] The maximum number of activity groups M_G is a required input to the algorithm but is not defined in the main text or given a default value. Its role in the absorption/ejection step should be stated.
  5. [Section 7] The paper discusses careful calibration of ε in the simulation study, but for the real-data analysis it does not report the value of ε used or the calibration procedure. This should be stated for reproducibility.
  6. [Section 3] The relation α = α_c + α_r for the conditional jump-length distribution is stated without derivation or citation; please provide a reference or a brief derivation.

Circularity Check

1 steps flagged · score 4.0 of 10

Jump-length agreement in Table 2 is an in-sample posterior fit, but independent checks give the model real content.

  1. fitted input called prediction [Section 5, 'Pure Jump Size Distribution'; Section 7, Table 2]
    "We model jump lengths as a Pareto distribution with common parameters α, ε between activity groups, that is, fΔr/Δt(Δr/Δt; α, ε)=αε^α(Δr/Δt)^{-(α+1)}1{(Δr/Δt)≥ε}. ... Table 2 displays the mean and standard deviation of jump length, MSD, and radius of gyration ... Whereas the LFCM closely reproduces jump length, MSD, and radius of gyration, simulations from the CPT estimator tend to exhibit under-bias on mean jump length and MSD."

    The LFCM likelihood in Eq. (23) defines jump lengths as Pareto random variables whose parameters (α, ε) are estimated from the same observed jump-length data (ε by the empirical KS/AD/Kuiper fitting described in Section 5; α through the MCMC posterior of Table 1). Consequently, when Table 2 reports that trajectories simulated from the MAP posterior have jump-length mean 1.580 versus the observed 1.556, the agreement is not an emergent prediction but a draw from the fitted Pareto likelihood that was itself fit to those observed jump lengths. The claim that the LFCM 'captures' jump length is therefore an in-sample posterior predictive check of an assumed distribution, not an independent validation. The MSD and radius-of-gyration comparisons retain more independent content because they depend

full rationale

The LFCM's central construction (Definition 5.1, Eq. 23) is an assumed generative mixture: jumps are modeled with a Pareto distribution, local movement with Brownian mixtures, and returns with Gaussian activity regions. This is a model assumption, not a derivation from the data. The main in-sample evidence is Table 2, where the same observations used to estimate Pareto parameters are then 'reproduced' by posterior simulation; as argued above, the jump-length agreement is partly forced by construction. However, several important checks are genuinely external or not fitted to the claimed outcome: the pooled tail index α ≈ 1.71–1.74 in Section 3 is compared with independent prior estimates (Brockmann et al., Gonzalez et al., Zhao et al.); the simulated rate of newly visited locations t^0.52 in Figure 19 is compared with Song et al.'s external t^0.6 benchmark; and the activity-space stability analysis in Section 7 is a self-contained convergence assessment. I found no load-bearing self-citation chain, no imported uniqueness theorem, and no renaming of a known result. The Lévy-flight-versus-Lévy-walk limitation is a substantive modeling concern but not an instance of circular reasoning: the paper explicitly acknowledges it as an assumption (Section 5). Overall, the central claim is not reducible to its inputs by construction, but one headline quantitative comparison is an in-sample fit, so a moderate circularity score is appropriate.

Assumptions & free parameters 3 free parameters · 8 assumptions · 0 invented entities

The central claim rests on several modeling assumptions: the Lévy flight approximation, the Pareto jump model, the constant jump probability, and the Gaussian return-probability form. The only data-fitted parameter of note is the minimal jump length ε, whose calibration is acknowledged as delicate. No new physical or conceptual entities are postulated; the activity regions are latent variables in a statistical model.

free parameters (3)
  • ε (minimal jump length) = 0.1 in simulation (visual inspection); unspecified in real-data analysis
    Threshold separating jumps from Brownian motion; Figure 15 shows results change substantially with ε; Section 6 advises fitting LFCM on a grid of ε values.
  • MG (maximum number of activity groups) = not stated
    Algorithm 2 requires an upper bound MG for the absorption/ejection sampler; no guidance given in the paper.
  • Neighborhood averaging radius r and neighborhood size k = not given
    Used to estimate ε in Section 5; no sensitivity analysis or default values provided.
assumptions (8)
  • standard math Lévy-Itô decomposition theorem
    Invoked in Section 5 to represent a Lévy process as Brownian motion plus compound Poisson plus martingale.
  • domain assumption Small jumps of the Lévy process are well approximated by Brownian motion
    Section 5 'Jump Approximation' uses Asmussen and Rosiński (2001) to replace the martingale term with Brownian motion for small ε; the paper then uses this approximation at a fixed, non-infinitesimal ε.
  • domain assumption The movement process is a Lévy flight, not a Lévy walk
    Explicit assumption in Section 5: jumps are instantaneous and independent of travel time, despite GPS data having timestamps.
  • domain assumption Jump lengths follow a Pareto distribution with common α and ε across activity groups
    The compound Poisson process in the decomposition is replaced by a Pareto 'data-driven simplification' rather than derived from the jump measure.
  • domain assumption Jump angles follow a von Mises distribution
    Conjugate prior choice for angular movement, Section 5.
  • ad hoc to paper Return to an activity region is distributed as N(μz, Σz) with the specific form in Eq. (22)
    The conditional return classification in Section 5.1 assumes a Gaussian form based on average Brownian movement over [0,T]; introduced for computational convenience, not derived from the movement process.
  • domain assumption Jump probability ν is constant and does not depend on Δt
    Eq. (17) defines ν = P(N(Δt)>0), but the prior model treats b ~ Bern(ν) with a single ν, ignoring the time scaling in the likelihood.
  • domain assumption First differences are independent conditional on latent states
    The model conditions on latent states and treats first differences as independent, a standard state-space assumption that ignores serial correlation in returns beyond the latent states.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Modeling Human Spatial Mobility Patterns with the L\'evy Flight Cluster Model." pith.science (2026). https://pith.science/paper/YANC2OCG

@misc{pith2026250900298,
  author       = {Pith},
  title        = {Pith review of: Modeling Human Spatial Mobility Patterns with the L\'evy Flight Cluster Model},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YANC2OCG}},
  note         = {Machine review of arXiv:2509.00298}
}
read the original abstract

Despite the extensive collection of individual mobility data over the past decade, fueled by the widespread use of GPS-enabled personal devices, the existing statistical literature on estimating human spatial mobility patterns from temporally irregular location data remains limited. In this paper, we introduce the L\'{e}vy Flight Cluster Model (LFCM), a hierarchical Bayesian mixture model designed to analyze an individual's activity distribution. The LFCM can be utilized to determine probabilistic overlaps between individuals' activity patterns and serves as an anonymization tool to generate synthetic location data. We present our methodology using real-world human location data, demonstrating its ability to accurately capture the key characteristics of human movement.

Figures

Figures reproduced from arXiv: 2509.00298 by the authors.

Figure 1
Figure 1. Distribution of records by day of week from November 2018 to Jan￾uary 2019. Each record includes a unique device identifier, geographic coordinates, a timestamp, an estimated margin of error for the coordinates, and a reported speed of travel. This allows iden￾tification of daily movement patterns for individual devices over the 12-week period [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Trajectory recorded for three individual devices over a one-week period starting on November 12, 2018. Latitude and longitude are shifted to maintain privacy. Numerous mobile device applications carry out location-sensitive data collection while in operation. This leads to a significant fluctuation in timestamps by hour of the day and day of the year [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Distribution of the number of records by hour of day for an example individual over the 12 week period from 11/01/18 to 01/30/19. 2008), mean squared displacement (Alessandretti et al., 2017), and frequent locations (Song et al., 2010). These metrics, applied within defined time periods, can capture unique aspects of human mobility. Many studies find that human travel often follows a heavy-tailed distribution —- wit… view at source ↗
Figures from the paper (22 more)
Figure 4
Figure 4. Figure 4: Histogram of jump lengths for 293 individuals in the analytic data set alongside the function f∆rp∆rq9p∆rq ´1.59 . between consecutive GPS locations within each (individual, week) group and analyze the upper tail of these lengths with a continuous power law model. For …
Figure 5
Figure 5. Figure 5: ECCDFs of jump distance on the log-log scale. Each plot is an individual device, and each curve is a week of observation. Mean Squared Displacement (MSD). Characteristically, MSD quantifies the average squared distance an entity moves from its initial position up to a …
Figure 6
Figure 6. Figure 6: Three instances of random walks each run for 7,000 steps. Brownian motion steps are shown in black and L´evy flight jumps are shown in red. Temporal Independence. The temporal dependence in Eq. (14) can be largely removed by considering the first difference representat…
Figure 7
Figure 7. Figure 7: Survival functions for normalized jump lengths ∆rt{∆t for three example individuals. The left panel depicts the entire distribution, whereas the panel figure depicts the zoomed in area denoted by the black box. We model jump lengths as a Pareto distribution with common…
Figure 8
Figure 8. Figure 8: Mean estimate of α and β for 5, 000 samples drawn from a Pareto distribution, with α “ 2.5, β “ 0.2, ε “ 100, and n “ 2, 500, plotted as a function for ε. Estimates are generated with power law and truncated power law distributions. 14 [PITH_FULL_IMAGE:figures/full_fi…
Figure 9
Figure 9. Figure 9: Left: CCDF of 10,000 samples generated by Eq. (20). Middle: Average fitted ε by method and number of samples. Right: average αpMLE by number of samples. Pure Jump Angle Distribution. A standard L´evy flight assumes a uniform angular distribution of movement, but in pra…
Figure 10
Figure 10. Figure 10: Radial histograms of jump angles for the individual devices in Fig￾ure 7 over the 12 week period. Due to its favorable conjugacy properties (Guttorp and Lockhart, 1988), a convenient choice to model angular movement is the von Mises distribution with density (21) fpϑ …
Figure 11
Figure 11. Figure 11: provides an illustrative example. Suppose that at some point in the past an individual was active in a region z with center µr z and covariance Σz, denoted by the red sample path in the figure. At time ti an individual moves to xptiq according to the jump distribution…
Figure 12
Figure 12. Figure 12: the entire set of simulated data, a random subset consisting of 50% of the simulated [PITH_FULL_IMAGE:figures/full_fig_p018_12.png]
Figure 12
Figure 12. Figure 12: Simulated GPS location data. subsets to estimate the activity distributions with LFCM together with the CPT estimator of Dong et al. (2020). The Conservative Proportional Time Estimator (CPT). The CPT estimator is a discrete esti￾mator of an individual’s activity dist…
Figure 13
Figure 13. Figure 13: Activity space estimates LFCM classification of simulated data (left) and the kernel density estimator (Chen and Dobra, 2020) is displayed as a reference (right). The subsets displayed are 100% (top), 50% (middle), and 25% (bottom) of the original data [PITH_FULL_IMA…
Figure 14
Figure 14. Figure 14: Proportional Time estimates with grid size 0.15 (left), 0.20 (mid￾dle), and 0.25 (right). three distinct grid sizes that substantively change the level set Ap0.9. From left to right, the grid sizes used are 0.15, 0.20, and 0.25. A grid size of 0.15 identifies addition…
Figure 15
Figure 15. Figure 15: LFCM with minimal jump distance parameter ε “ 0.01 (left panel), 0.10 (middle panel), and 0.20 (right panel). Persistence Curves. The activity regions estimated by the LFCM naturally provide a probability density from the mixture of Brownian motions which generate the…
Figure 16
Figure 16. Figure 16: Persistence curve of activity regions for the data (black), CPT (yellow), and LFCM (red). of a connected component is defined by its birth and death times, which are determined by the corresponding level of density ranking. A decrease in the number of connected compon…
Figure 17
Figure 17. Figure 17: Example of LFCM on 4 weeks of human mobility data. Colored points represent distinct activities, gray points represent jumps, and shaded areas represent activity regions, according to the LFCM estimate. a location record identified as a non-Brownian jump in the observ…
Figure 18
Figure 18. Figure 18: Simulated movement for an individual device based on the LFCM MAP estimate. This simulation process also results in reasonable mobility metrics mentioned in Section 3 [PITH_FULL_IMAGE:figures/full_fig_p023_18.png]
Figure 19
Figure 19. Figure 19: Representative figure for the LFCM simulated proportion of new locations visited over time for six simulated devices (colored) relative to fptq “ t 0.6 (dashed), the proportion of new locations visited estimated in (Song et al., 2010). Activity Space Stability. While …
Figure 20
Figure 20. Figure 20: LFCM-based activity space distributions for an individual device after 1 week (left), 1 month (middle), and 3 months (right) [PITH_FULL_IMAGE:figures/full_fig_p025_20.png]
Figure 21
Figure 21. Figure 21: “Top 1” activity area (left),“Top 3” activity areas (middle), and the 80% activity region (right), estimated using the LFCM with w P t2, 4, . . . , 12u weeks of data. Finally, to construct a quantitative measure of convergence for the LFCM activity space es￾timates, w…
Figure 22
Figure 22. Figure 22: Mean proportion overlap averaged across devices for “Top 1” ac￾tivity area (left), “Top 3” activity areas (middle), and 80% activity area (right). Probabilistic Contact Networks. In social and behavioral sciences, understanding the proba￾bilistic contact networks can …
Figure 23
Figure 23. Figure 23: Example interpolated and extrapolated trajectory for individual de￾vice trajectory across one week. Latitude and longitude have been anonymized. from the target posterior distribution ppθ | Dq, after convergence is achieved [PITH_FULL_IMAGE:figures/full_fig_p027_23.png]
Figure 24
Figure 24. Figure 24: Mean pairwise distance matrices estimated using LFCM for interpo￾lation/extrapolation across 500 generated sample paths (left), and linear inter￾polation/extrapolation for the same week (right). long-range movements consistent with L´evy flight dynamics. These dual co…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Bayesian Gaussian Mixture Modeling for Symmetric Matrix Variate Data

    stat.ME 2026-08 conditional novelty 6.0 of 10

    A Bayesian mixture model for symmetric matrix data writes one precision matrix as a single power of the other; applied to GPS mobility, it links crime and youth employment to weekly activity patterns.

Reference graph

Works this paper leans on

65 extracted references · 59 canonical work pages · cited by 1 Pith paper

  1. [1]

    Handbook of mathematical functions with formulas, graphs, and mathematical table

    Milton Abramowitz and Irene A Stegun. Handbook of mathematical functions with formulas, graphs, and mathematical table. US Department of Commerce; National Bureau of Standards Applied Mathematics Series, 55, 1965

    work page 1965

  2. [2]

    Boone, and Grace S

    Mahshid Ahmadian, Edward L. Boone, and Grace S. Chiu. A new perspective to fish trajectory imputation: A methodology for spatiotemporal modeling of acoustically tagged fish data, 2025

    work page 2025

  3. [3]

    Multi-scale spatio-temporal analysis of human mobility

    Laura Alessandretti, Piotr Sapiezynski, Sune Lehmann, and Andrea Baronchelli. Multi-scale spatio-temporal analysis of human mobility. PloS One, 12 0 (2): 0 e0171686, 2017

    work page 2017

  4. [4]

    Linking movement ecology with wildlife management and conservation

    Andrew M Allen and Navinder J Singh. Linking movement ecology with wildlife management and conservation. Frontiers in Ecology and Evolution, 3: 0 155, 2016

    work page 2016

  5. [5]

    Barry C. Arnold. Pareto Distribution, pages 1--10. John Wiley & Sons, Ltd, 2015. ISBN 9781118445112. doi:https://doi.org/10.1002/9781118445112.stat01100.pub2. URL https://onlinelibrary.wiley.com/doi/abs/10.1002/9781118445112.stat01100.pub2

    doi:10.1002/9781118445112.stat01100.pub2 2015

  6. [6]

    Approximations of small jumps of l \'e vy processes with a view towards simulation

    S ren Asmussen and Jan Rosi \'n ski. Approximations of small jumps of l \'e vy processes with a view towards simulation. Journal of Applied Probability, 38 0 (2): 0 482--493, 2001

    work page 2001

  7. [7]

    Eugene Stanley

    Albert-L \'a szl \'o Barab \'a si and H. Eugene Stanley. Fractal Concepts in Surface Growth. Cambridge University Press, 1995

    work page 1995

  8. [8]

    L \'e vy processes , volume 121

    Jean Bertoin. L \'e vy processes , volume 121. Cambridge university press Cambridge, 1996

    work page 1996

Show all 65 references
  1. [9]

    Deriving and validating trip purposes and travel modes for multi-day gps-based travel surveys: A large-scale application in the netherlands

    Wendy Bohte and Kees Maat. Deriving and validating trip purposes and travel modes for multi-day gps-based travel surveys: A large-scale application in the netherlands. Transportation Research Part C: Emerging Technologies, 17 0 (3): 0 285--297, 2009

    2009

  2. [10]

    The scaling laws of human travel

    Dirk Brockmann, Lars Hufnagel, and Theo Geisel. The scaling laws of human travel. Nature, 439 0 (7075): 0 462--465, 2006

    2006

  3. [11]

    Urban form and household activity-travel behavior

    Ron N Buliung and Pavlos S Kanaroglou. Urban form and household activity-travel behavior. Growth and Change, 37 0 (2): 0 172--199, 2006

    2006

  4. [12]

    Exploring spatial variety in patterns of activity-travel behaviour: initial results from the toronto travel-activity panel survey (ttaps)

    Ron N Buliung, Matthew J Roorda, and Tarmo K Remmel. Exploring spatial variety in patterns of activity-travel behaviour: initial results from the toronto travel-activity panel survey (ttaps). Transportation, 35 0 (6): 0 697--722, 2008

    2008

  5. [13]

    The fractal geometry of nature

    JW Cannon. The fractal geometry of nature. by benoit b. mandelbrot. The American Mathematical Monthly, 91 0 (9): 0 594--598, 1984

    1984

  6. [14]

    Measuring human activity spaces from gps data with density ranking and summary curves

    Yen-Chi Chen and Adrian Dobra. Measuring human activity spaces from gps data with density ranking and summary curves. Annals of Applied Statistics, 14 0 (1): 0 409--432, 2020

    2020

  7. [15]

    Power-law distributions in empirical data

    Aaron Clauset, Cosma Rohilla Shalizi, and Mark EJ Newman. Power-law distributions in empirical data. SIAM review, 51 0 (4): 0 661--703, 2009

    2009

  8. [16]

    Spatiotemporal detection of unusual human population behavior using mobile phone data

    Adrian Dobra, Nathalie E Williams, and Nathan Eagle. Spatiotemporal detection of unusual human population behavior using mobile phone data. PloS One, 10 0 (3): 0 e0120449, 2015

    2015

  9. [17]

    A statistical framework for measuring the temporal stability of human mobility patterns

    Zhihang Dong, Yen-Chi Chen, and Adrian Dobra. A statistical framework for measuring the temporal stability of human mobility patterns. Journal of Applied Statistics, pages 1--19, 2020

    2020

  10. [18]

    Stochastic processes, volume 7

    Joseph L Doob and Joseph L Doob. Stochastic processes, volume 7. Wiley New York, 1953

    1953

  11. [19]

    Comparing distributions: the two-sample anderson-darling test as an alternative to the kolmogorov-smirnoff test

    Sonja Engmann and Denis Cousineau. Comparing distributions: the two-sample anderson-darling test as an alternative to the kolmogorov-smirnoff test. Journal of applied quantitative methods, 6 0 (3), 2011

    2011

  12. [20]

    Urban form, individual spatial footprints, and travel: Examination of space-use behavior

    Yingling Fan and Asad J Khattak. Urban form, individual spatial footprints, and travel: Examination of space-use behavior. Transportation Research Record, 2082 0 (1): 0 98--106, 2008

  13. [21]

    An Introduction to Probability Theory and Its Applications, Volume II

    William Feller. An Introduction to Probability Theory and Its Applications, Volume II. John Wiley & Sons, 1971

    1971

  14. [22]

    Closure of ‘third places’? exploring potential consequences for collective health and wellbeing

    Jessica Finlay, Michael Esposito, Min Hee Kim, Iris Gomez-Lopez, and Philippa Clarke. Closure of ‘third places’? exploring potential consequences for collective health and wellbeing. Health & place, 60: 0 102225, 2019

    2019

  15. [23]

    Gardiner

    Crispin W. Gardiner. Stochastic Methods: A Handbook for the Natural and Social Sciences. Springer, 4th edition, 2009

    2009

  16. [24]

    anomalous

    Stefan Gheorghiu and Marc-Olivier Coppens. Heterogeneity explains features of “anomalous” thermodynamics and statistics. Proceedings of the National Academy of Sciences, 101 0 (45): 0 15852--15856, 2004

    2004

  17. [25]

    Limit distributions for sums of independent

    BV Gnedenko, AN Kolmogorov, BV Gnedenko, and AN Kolmogorov. Limit distributions for sums of independent. Am. J. Math, 105, 1954

    1954

  18. [26]

    Crossing the border? exploring the cross-state mobility of the teacher workforce

    Dan Goldhaber, Cyrus Grout, Kristian L Holden, and Nate Brown. Crossing the border? exploring the cross-state mobility of the teacher workforce. Educational Researcher, 44 0 (8): 0 421--431, 2015

    2015

  19. [27]

    Understanding individual human mobility patterns

    Marta C Gonzalez, Cesar A Hidalgo, and Albert-Laszlo Barabasi. Understanding individual human mobility patterns. Nature, 453 0 (7196): 0 779--782, 2008

    2008

  20. [28]

    Finding the location of a signal: A bayesian analysis

    Peter Guttorp and Richard A Lockhart. Finding the location of a signal: A bayesian analysis. Journal of the American Statistical Association, 83 0 (402): 0 322--330, 1988

    1988

  21. [29]

    Development of M o A L s ( M obile A ctivity L oggers supported by G S phones) for travel behavior analysis

    Eiji Hato. Development of M o A L s ( M obile A ctivity L oggers supported by G S phones) for travel behavior analysis. In Proceedings of the 85th Annual Meeting of the Transportation Research Board. Washington D.C., 2006

    2006

  22. [30]

    Hooten, Devin S

    Mevin B. Hooten, Devin S. Johnson, Brett T. McClintock, and Juan M. Morales. Animal movement: statistical models for telemetry data. CRC press, 2017

    2017

  23. [31]

    Analyzing animal movements using brownian bridges

    Jon S Horne, Edward O Garton, Stephen M Krone, and Jesse S Lewis. Analyzing animal movements using brownian bridges. Ecology, 88 0 (9): 0 2354--2363, 2007

    2007

  24. [32]

    On stochastic processes (i) infinitely divisible laws of probability

    Kiyosi It \^o . On stochastic processes (i) infinitely divisible laws of probability. In Japanese journal of mathematics: transactions and abstracts, volume 18, pages 261--301. The Mathematical Society of Japan, 1941

    1941

  25. [33]

    Terrestrial animal tracking as an eye on life and planet

    Roland Kays, Margaret C Crofoot, Walter Jetz, and Martin Wikelski. Terrestrial animal tracking as an eye on life and planet. Science, 348 0 (6240): 0 aaa2478, 2015

    2015

  26. [34]

    Transport and social exclusion: investigating the possibility of promoting inclusion through virtual mobility

    Susan Kenyon, Glenn Lyons, and Jackie Rafferty. Transport and social exclusion: investigating the possibility of promoting inclusion through virtual mobility. Journal of Transport Geography, 10 0 (3): 0 207--219, 2002

    2002

  27. [35]

    LaPoint, Martin Wikelski, and Kamran Safi

    Bart Kranstauber, Roland Kays, Scott D. LaPoint, Martin Wikelski, and Kamran Safi. A dynamic brownian bridge movement model to estimate utilization distributions for heterogeneous animal movement. Journal of Animal Ecology, 81: 0 738--746, 2012

    2012

  28. [36]

    Bivariate gaussian bridges: directional factorization of diffusion in brownian bridge models

    Bart Kranstauber, Kamran Safi, and Frederic Bartumeus. Bivariate gaussian bridges: directional factorization of diffusion in brownian bridge models. Movement Ecology, 2 0 (1): 0 1--10, 2014

    2014

  29. [37]

    Brownian bridge interpolation for human mobility? In Proceedings of the 29th International Conference on Advances in Geographic Information Systems, pages 175--183, 2021

    John Krumm. Brownian bridge interpolation for human mobility? In Proceedings of the 29th International Conference on Advances in Geographic Information Systems, pages 175--183, 2021

    2021

  30. [38]

    The uncertain geographic context problem

    Mei-Po Kwan. The uncertain geographic context problem. Annals of the Association of American Geographers, 102 0 (5): 0 958--968, 2012

    2012

  31. [39]

    Th \'e orie de l’addition des variables al \'e atoires

    P L \'e vy. Th \'e orie de l’addition des variables al \'e atoires. Bull. Amer. Math. Soc, 44 0 (Part 1): 0 19--20, 1938

    1938

  32. [40]

    Temporal understanding of human mobility: A multi-time scale analysis

    Tongtong Liu, Zheng Yang, Yi Zhao, Chenshu Wu, Zimu Zhou, and Yunhao Liu. Temporal understanding of human mobility: A multi-time scale analysis. PloS One, 13 0 (11): 0 e0207697, 2018

    2018

  33. [41]

    Explore spatiotemporal and demographic characteristics of human mobility via twitter: A case study of chicago

    Feixiong Luo, Guofeng Cao, Kevin Mulligan, and Xiang Li. Explore spatiotemporal and demographic characteristics of human mobility via twitter: A case study of chicago. Applied Geography, 70: 0 11--25, 2016

    2016

  34. [42]

    Mandelbrot and John W

    Benoit B. Mandelbrot and John W. Van Ness. Fractional brownian motions, fractional noises and applications. SIAM Review, 10 0 (4): 0 422--437, 1968

    1968

  35. [43]

    Spatial polygamy and contextual exposures (spaces) promoting activity space approaches in research on place and health

    Stephen A Matthews and Tse-Chuan Yang. Spatial polygamy and contextual exposures (spaces) promoting activity space approaches in research on place and health. American behavioral scientist, 57 0 (8): 0 1057--1081, 2013

    2013

  36. [44]

    The ecology of human mobility

    Mark G Meekan, Carlos M Duarte, Juan Fern \'a ndez-Gracia, Michele Thums, Ana MM Sequeira, Rob Harcourt, and V \' ctor M Egu \' luz. The ecology of human mobility. Trends in ecology & evolution, 32 0 (3): 0 198--210, 2017

    2017

  37. [45]

    Urban activity spaces: Illustrations and application of a conceptual model for integrating the time and space dimensions

    Tracy H Newsome, Wayne A Walcott, and Paul D Smith. Urban activity spaces: Illustrations and application of a conceptual model for integrating the time and space dimensions. Transportation, 25 0 (4): 0 357--377, 1998

    1998

  38. [46]

    Bayesian finite mixtures with an unknown number of components: The allocation sampler

    Agostino Nobile and Alastair T Fearnside. Bayesian finite mixtures with an unknown number of components: The allocation sampler. Statistics and Computing, 17 0 (2): 0 147--162, 2007

    2007

  39. [47]

    The third place

    Ramon Oldenburg and Dennis Brissett. The third place. Qualitative sociology, 5 0 (4): 0 265--284, 1982

    1982

  40. [48]

    The problem of the random walk

    Karl Pearson. The problem of the random walk. Nature, 72 0 (1865): 0 294--294, 1905

    1905

  41. [49]

    Pandemic precarity: Covid-19 is exposing and exacerbating inequalities in the american heartland

    Brea L Perry, Brian Aronson, and Bernice A Pescosolido. Pandemic precarity: Covid-19 is exposing and exacerbating inequalities in the american heartland. Proceedings of the National Academy of Sciences, 118 0 (8), 2021

    2021

  42. [50]

    Injong Rhee, Minsu Shin, Seongik Hong, Kyunghan Lee, S. J. Kim, and Song Chong. On the levy-walk nature of human mobility. IEEE/ACM transactions on networking, 19 0 (3): 0 630--643, 2011

    2011

  43. [51]

    On bayesian analysis of mixtures with an unknown number of components (with discussion)

    Sylvia Richardson and Peter J Green. On bayesian analysis of mixtures with an unknown number of components (with discussion). Journal of the Royal Statistical Society: series B (statistical methodology), 59 0 (4): 0 731--792, 1997

    1997

  44. [52]

    Plasticity in elk migration timing is a response to changing environmental conditions

    Gregory JM Rickbeil, Jerod A Merkle, Greg Anderson, M Paul Atwood, Jon P Beckmann, Eric K Cole, Alyson B Courtemanch, Sarah Dewey, David D Gustine, Matthew J Kauffman, et al. Plasticity in elk migration timing is a response to changing environmental conditions. Global change b...

    2019

  45. [53]

    Bayesian model selection for the latent position cluster model for social networks

    Caitriona Ryan, Jason Wyse, and Nial Friel. Bayesian model selection for the latent position cluster model for social networks. Network Science, 5: 0 70--91, 03 2017. doi:10.1017/nws.2017.6

    2017 doi

  46. [54]

    Scharf, Mevin B

    Henry R. Scharf, Mevin B. Hooten, Timothy J. Hefley, and David E. Russell. Dynamic social networks inferred from movement data via latent regime switching. Journal of Animal Ecology, 84 0 (5): 0 1230--1241, 2015

    2015

  47. [55]

    Scharf, Timothy Hefley, Daniel Morin, and Mevin B

    Henry R. Scharf, Timothy Hefley, Daniel Morin, and Mevin B. Hooten. Hierarchical bayesian imputation of animal trajectories from sparse telemetry data. Journal of Wildlife Management, 81 0 (2): 0 123--136, 2017

    2017

  48. [56]

    Activity spaces: measures of social exclusion? Transport policy, 10 0 (4): 0 273--286, 2003

    Stefan Sch \"o nfelder and Kay W Axhausen. Activity spaces: measures of social exclusion? Transport policy, 10 0 (4): 0 273--286, 2003

    2003

  49. [57]

    On the variability of human activity spaces

    Stefan Sch \"o nfelder and Kay W Axhausen. On the variability of human activity spaces. In The Real and Virtual Worlds of Spatial Planning, pages 237--262. Springer, 2004

    2004

  50. [58]

    A suite of methods for representing activity space in a healthcare accessibility study

    Jill E Sherman, John Spencer, John S Preisser, Wilbert M Gesler, and Thomas A Arcury. A suite of methods for representing activity space in a healthcare accessibility study. International journal of health geographics, 4 0 (1): 0 1--21, 2005

    2005

  51. [59]

    Modelling the scaling properties of human mobility

    Chaoming Song, Tal Koren, Pu Wang, and Albert-L \'a szl \'o Barab \'a si. Modelling the scaling properties of human mobility. Nature Physics, 6 0 (10): 0 818--823, 2010

    2010

  52. [60]

    Deducing mode and purpose from gps data

    Peter Stopher, Eoin Clifford, Jun Zhang, and Camden FitzGerald. Deducing mode and purpose from gps data. 2008. URL http://hdl.handle.net/2123/19552

    2008

  53. [61]

    Van Kampen

    N.G. Van Kampen. Stochastic Processes in Physics and Chemistry. Elsevier, 3rd edition, 2007

    2007

  54. [62]

    Measures of human mobility using mobile phone records enhanced with gis data

    Nathalie E Williams, Timothy A Thomas, Matthew Dunbar, Nathan Eagle, and Adrian Dobra. Measures of human mobility using mobile phone records enhanced with gis data. PloS One, 10 0 (7): 0 e0133630, 2015

    2015

  55. [63]

    Elimination of the travel diary: Experiment to derive trip purpose from global positioning system travel data

    Jean Wolf, Randall Guensler, and William Bachman. Elimination of the travel diary: Experiment to derive trip purpose from global positioning system travel data. Transportation Research Record, 1768 0 (1): 0 125--134, 2001

    2001

  56. [64]

    Block clustering with collapsed latent block models

    Jason Wyse and Nial Friel. Block clustering with collapsed latent block models. Statistics and Computing, 22 0 (2): 0 415--428, 2012

    2012

  57. [65]

    Explaining the power-law distribution of human mobility through transportationmodality decomposition

    Kai Zhao, Mirco Musolesi, Pan Hui, Weixiong Rao, and Sasu Tarkoma. Explaining the power-law distribution of human mobility through transportationmodality decomposition. Scientific reports, 5 0 (1): 0 1--7, 2015

    2015

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.