pith. sign in

arxiv: 2605.12830 · v2 · pith:JWSX5JW4new · submitted 2026-05-12 · 📊 stat.ME

Linking COPD Prevalence with Income Distribution: A Spatial Heterogeneous Compositional Regression via Geographically Weighted Penalized Approach

Jingwen Deng , Shujie Ma , Sergio J. Rey , Guanyu Hu This is my paper

Pith reviewed 2026-05-19 17:08 UTC · model grok-4.3

classification 📊 stat.ME
keywords spatial regressioncompositional datafusion penaltyCOPD prevalenceincome distributiongeographically weightedspatial heterogeneityhealth disparities
0
0 comments X

The pith

A new regression model with pairwise fusion penalties detects both adjacent and non-adjacent regions sharing the same income-to-COPD links.

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

This paper develops a geographically weighted compositional regression that adds a pairwise fusion penalty to group regions with identical regression coefficients linking income bracket proportions to COPD prevalence rates. The penalty works without assuming that regions must be neighbors or that effects must change smoothly across space. Nonconvex penalties such as the minimax concave penalty are included to support stable estimation when many income categories and spatial units are present. In the U.S. county application, the model uncovers clusters of similar associations that conventional spatial smoothers average away or cannot separate. The result is a practical tool for mapping how compositional income data relate to health outcomes under realistic geographic heterogeneity.

Core claim

The paper claims that a geographically weighted penalized compositional regression equipped with a pairwise fusion penalty can identify clusters of regions—contiguous or not—that share the same regression effects between income distributions and COPD prevalence, while using nonconvex penalties to maintain estimation accuracy and interpretability in high-dimensional spatial settings.

What carries the argument

The pairwise fusion penalty, which shrinks differences in regression coefficients across all pairs of regions to form clusters with shared effects regardless of geographic adjacency.

If this is right

  • Regions with comparable income structures can be grouped for joint analysis even when they do not share a border.
  • Abrupt spatial changes in how income affects COPD rates become detectable without forcing gradual transitions.
  • High-dimensional compositional predictors can be processed with greater numerical stability and clearer cluster output.
  • Health-policy maps can highlight groups of similar areas rather than treating every location as unique.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same penalty structure could be tested on other health endpoints such as diabetes or mental-health measures using analogous compositional covariates.
  • Identified clusters might correspond to known economic corridors or migration patterns that cross state lines.
  • Future work could replace the MCP with alternative nonconvex penalties and compare cluster recovery rates on the same COPD data.

Load-bearing premise

The pairwise fusion penalty combined with nonconvex penalties such as MCP will correctly recover noncontiguous clusters and deliver improved accuracy and scalability in high-dimensional spatial compositional data.

What would settle it

A controlled simulation with known noncontiguous clusters where the method recovers fewer or different clusters than the true grouping, or fails to show lower estimation error than standard geographically weighted regression.

Figures

Figures reproduced from arXiv: 2605.12830 by Guanyu Hu, Jingwen Deng, Sergio J. Rey, Shujie Ma.

Figure 1
Figure 1. Figure 1: Estimated COPD prevalence across the United States by state (left) and within Texas by [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Spatial weight patterns under different method, with r=8 for (b) and (c) [PITH_FULL_IMAGE:figures/full_fig_p012_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Spatial partitions at different geographic resolutions. The upper figure shows the state [PITH_FULL_IMAGE:figures/full_fig_p015_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Spatial clustering results under different values of the decay parameter [PITH_FULL_IMAGE:figures/full_fig_p019_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: State-level clustering results (left) and BIC comparison (right) under different spatial [PITH_FULL_IMAGE:figures/full_fig_p021_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: County-level clustering results (left) and BIC comparison (right) under different spatial [PITH_FULL_IMAGE:figures/full_fig_p023_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: lattice-based partition design In addition to the two simulation in main content, we include this lattice-based spatial structure to provide a controlled experimental setting in which spatial relationships are explicitly defined through regular adjacency. A key advantage of this design is that it allows the number of spatial units to be flexibly scaled, enabling systematic evaluation of model performance a… view at source ↗
read the original abstract

Income inequality is a major contributor to health disparities, yet its effects often vary by geography and are commonly represented as compositional distributions (e.g., proportions of households across income brackets). Existing spatial regression methods struggle in this setting: they typically assume smooth spatial variation, cannot accommodate abrupt spatial heterogeneity, and lack principled treatment of compositional covariates. We propose a geographically weighted penalized compositional regression model that addresses these challenges simultaneously. Our method adopts a pairwise fusion penalty that enables detection of both contiguous and noncontiguous regional clusters with shared regression effects, thereby relaxing strong assumptions of spatial smoothness and geographic contiguity. This allows regions with similar underlying socioeconomic structures to be identified even when they are not geographically adjacent. By incorporating nonconvex penalties, such as the minimax concave penalty (MCP), the approach achieves improved estimation accuracy, interpretability, and scalability in high-dimensional spatial settings. We illustrate the method through an analysis linking U.S. income composition to chronic obstructive pulmonary disease (COPD) prevalence, revealing spatially heterogeneous associations that are obscured by conventional models. The proposed framework provides a flexible and robust tool for spatial data analysis involving compositional predictors and region-specific heterogeneity.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript proposes a geographically weighted penalized compositional regression model that uses a pairwise fusion penalty (combined with nonconvex penalties such as MCP) to detect both contiguous and noncontiguous regional clusters sharing regression effects. It relaxes assumptions of spatial smoothness and geographic contiguity in the analysis of compositional predictors, and applies the method to link U.S. income-bracket proportions to COPD prevalence, claiming to uncover spatially heterogeneous associations missed by standard models.

Significance. If the fusion penalty demonstrably overrides the geographic kernel to recover noncontiguous clusters without introducing indirect contiguity bias, the framework would offer a useful advance for spatial compositional regression in health-disparities research, providing greater flexibility than conventional GWR or fused-lasso spatial models while maintaining interpretability through cluster detection.

major comments (2)
  1. [Abstract (proposed model paragraph)] Abstract (paragraph on the proposed model): the claim that the pairwise fusion penalty 'enables detection of both contiguous and noncontiguous regional clusters' and 'relaxes strong assumptions of spatial smoothness and geographic contiguity' is not accompanied by any derivation or argument showing that the fusion term can dominate the distance-based local loss for non-adjacent regions; the kernel bandwidth and fusion tuning parameter necessarily interact, so it remains possible that strong geographic weighting still penalizes noncontiguous grouping indirectly.
  2. [Model description] Model description: no explicit objective function, loss term for compositional covariates, or optimization procedure is supplied, making it impossible to verify that the joint estimator separates the effects of the geographically weighted kernel from the pairwise fusion penalty as required for the central noncontiguity claim.
minor comments (1)
  1. The abstract would be strengthened by a single-line statement of the objective function or the form of the fusion penalty to allow readers to assess the claimed separation of effects without waiting for the full methods section.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback on our manuscript. We address each major comment below, indicating planned revisions where appropriate to improve clarity and rigor.

read point-by-point responses

  1. Referee: Abstract (proposed model paragraph): the claim that the pairwise fusion penalty 'enables detection of both contiguous and noncontiguous regional clusters' and 'relaxes strong assumptions of spatial smoothness and geographic contiguity' is not accompanied by any derivation or argument showing that the fusion term can dominate the distance-based local loss for non-adjacent regions; the kernel bandwidth and fusion tuning parameter necessarily interact, so it remains possible that strong geographic weighting still penalizes noncontiguous grouping indirectly.

    Authors: We acknowledge the referee's concern that the abstract asserts noncontiguous cluster detection without an accompanying argument on how the fusion penalty interacts with the geographic kernel. In the proposed model the pairwise fusion penalty operates globally on coefficient differences across all region pairs irrespective of distance, while the kernel enters only through the local loss weights. This separation in principle allows sufficiently strong fusion to induce non-adjacent groupings. To address the gap we will add a short subsection in the methodology that derives the conditions under which the fusion term can dominate the kernel weighting and will include a small simulation illustration. The abstract will be revised to reference this new material. revision: yes

  2. Referee: Model description: no explicit objective function, loss term for compositional covariates, or optimization procedure is supplied, making it impossible to verify that the joint estimator separates the effects of the geographically weighted kernel from the pairwise fusion penalty as required for the central noncontiguity claim.

    Authors: We agree that the manuscript should have presented the explicit objective function. This omission prevents direct verification of the separation between the geographically weighted loss and the global fusion penalty. In the revision we will insert the full objective function, specify the compositional loss (via isometric log-ratio transformation of the income proportions), and describe the optimization routine (a block coordinate descent procedure with proximal mapping for the MCP and fusion terms). These additions will make the claimed separation transparent. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected; derivation is a self-contained methodological proposal

full rationale

The paper introduces a geographically weighted penalized compositional regression model that combines local kernel weighting with a pairwise fusion penalty (and nonconvex penalties such as MCP) to detect both contiguous and noncontiguous clusters. The abstract and described framework present this as a new modeling approach for handling spatial heterogeneity in compositional predictors, with the COPD-income application serving as illustration. No quoted derivation step reduces a claimed prediction, uniqueness result, or first-principles outcome to a fitted parameter or self-citation by construction. The central contribution is the joint objective and its claimed relaxation of smoothness/contiguity assumptions, which does not exhibit self-definitional, fitted-input, or load-bearing self-citation patterns in the provided text. The derivation chain remains independent of its own outputs.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The proposal rests on standard mathematical properties of penalized regression and compositional data transformations plus domain assumptions about spatial clustering; no new physical entities are introduced.

free parameters (1)
  • fusion penalty tuning parameter
    Controls cluster detection strength; value and selection method not specified in abstract.
axioms (1)
  • domain assumption Pairwise fusion penalty can identify noncontiguous clusters sharing regression effects
    Invoked to relax contiguity and smoothness requirements in the model description.

pith-pipeline@v0.9.0 · 5743 in / 1293 out tokens · 71511 ms · 2026-05-19T17:08:41.240720+00:00 · methodology

Review history (2 revisions) →

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

75 extracted references · 75 canonical work pages · 3 internal anchors

  1. [1]

    Public Health Regions , year =

    work page

  2. [2]

    Pleasants, R. A. and Riley, I. L. and Mannino, D. M. , title =. Int J Chron Obstruct Pulmon Dis , year =

    work page

  3. [3]

    and Siddharthan, T

    Grigsby, M. and Siddharthan, T. and Chowdhury, M. A. and Siddiquee, A. and Rubinstein, A. and Sobrino, E. and Miranda, J. J. and Bernabe-Ortiz, A. and Alam, D. and Checkley, W. , journal=. Socioeconomic status and. 2016 , month=. doi:10.2147/COPD.S111145 , pmid=

    work page doi:10.2147/copd.s111145 2016

  4. [4]

    2024 , journal =

    Endogenous Spatial Regimes , author =. 2024 , journal =. doi:10.1007/s10109-023-00411-2 , langid =

    work page doi:10.1007/s10109-023-00411-2 2024

  5. [5]

    Journal of the American Statistical Association , year=

    Spatial homogeneity pursuit of regression coefficients for large datasets , author=. Journal of the American Statistical Association , year=

    work page

  6. [6]

    Environmetrics , volume=

    Scanner: Simultaneously temporal trend and spatial cluster detection for spatial-temporal data , author=. Environmetrics , volume=. 2024 , publisher=

    work page 2024

  7. [7]

    Statistics in medicine , volume=

    Cluster detection of spatial regression coefficients , author=. Statistics in medicine , volume=. 2017 , publisher=

    work page 2017

  8. [8]

    Biometrika , volume=

    Variable selection in regression with compositional covariates , author=. Biometrika , volume=. 2014 , publisher=

    work page 2014

  9. [9]

    Spatial and Spatio-temporal Epidemiology , volume=

    Regularized spatial and spatio-temporal cluster detection , author=. Spatial and Spatio-temporal Epidemiology , volume=. 2022 , publisher=

    work page 2022

  10. [10]

    The International Journal of Biostatistics , author =

    Exploration of. The International Journal of Biostatistics , author =. 2020 , pages =. doi:10.1515/ijb-2018-0026 , abstract =

    work page doi:10.1515/ijb-2018-0026 2020

  11. [11]

    Statistics in Medicine , author =

    Multivariate log-contrast regression with sub-compositional predictors:. Statistics in Medicine , author =. 2022 , note =. doi:10.1002/sim.9273 , abstract =

    work page doi:10.1002/sim.9273 2022

  12. [12]

    Biometrics , author =

    It's all relative:. Biometrics , author =. 2023 , note =. doi:10.1111/biom.13703 , abstract =

    work page doi:10.1111/biom.13703 2023

  13. [13]

    The statistical analysis of compositional data.Journal of the Royal Statistical Society: Series B (Methodological), 44(2):139–160, 1982

    The. Journal of the Royal Statistical Society: Series B (Methodological) , author =. 1982 , pages =. doi:10.1111/j.2517-6161.1982.tb01195.x , abstract =

    work page doi:10.1111/j.2517-6161.1982.tb01195.x 1982

  14. [14]

    Distributed optimization and statistical learning via the alternating direction method of multipliers,

    Distributed. Foundations and Trends® in Machine Learning , author =. 2010 , pages =. doi:10.1561/2200000016 , language =

    work page doi:10.1561/2200000016 2010

  15. [15]

    Geographical Analysis , author =

    Geographically. Geographical Analysis , author =. 1996 , note =. doi:10.1111/j.1538-4632.1996.tb00936.x , abstract =

    work page doi:10.1111/j.1538-4632.1996.tb00936.x 1996

  16. [16]

    Journal of the American Statistical Association , author =

    Spatial. Journal of the American Statistical Association , author =. 2003 , pmid =. doi:10.1198/016214503000170 , abstract =

    work page doi:10.1198/016214503000170 2003

  17. [17]

    Journal of Econometrics , author =

    Shrinkage estimation of common breaks in panel data models via adaptive group fused. Journal of Econometrics , author =. 2016 , keywords =. doi:10.1016/j.jeconom.2015.09.004 , abstract =

    work page doi:10.1016/j.jeconom.2015.09.004 2016

  18. [18]

    Biometrics , author =

    Semiparametric. Biometrics , author =. 2010 , note =. doi:10.1111/j.1541-0420.2009.01309.x , abstract =

    work page doi:10.1111/j.1541-0420.2009.01309.x 2010

  19. [19]

    Biometrics , author =

    Bayesian. Biometrics , author =. 2010 , pages =. doi:10.1111/j.1541-0420.2009.01333.x , abstract =

    work page doi:10.1111/j.1541-0420.2009.01333.x 2010

  20. [20]

    Applied Physiology, Nutrition, and Metabolism , author =

    A systematic review of compositional data analysis studies examining associations between sleep, sedentary behaviour, and physical activity with health outcomes in adults , volume =. Applied Physiology, Nutrition, and Metabolism , author =. 2020 , note =. doi:10.1139/apnm-2020-0160 , abstract =

    work page doi:10.1139/apnm-2020-0160 2020

  21. [21]

    American Economic Review , author =

    Increasing. American Economic Review , author =. 2006 , pages =. doi:10.1257/aer.96.3.461 , abstract =

    work page doi:10.1257/aer.96.3.461 2006

  22. [22]

    Science of The Total Environment , author =

    Univariate statistical analysis of environmental (compositional) data:. Science of The Total Environment , author =. 2009 , keywords =. doi:10.1016/j.scitotenv.2009.08.008 , abstract =

    work page doi:10.1016/j.scitotenv.2009.08.008 2009

  23. [23]

    Nutrition Journal , author =

    A review of statistical methods for dietary pattern analysis , volume =. Nutrition Journal , author =. 2021 , keywords =. doi:10.1186/s12937-021-00692-7 , abstract =

    work page doi:10.1186/s12937-021-00692-7 2021

  24. [24]

    Bayesian

    Meng, Jingcheng and Ren, Yimeng and Zhu, Xuening and Hu, Guanyu , month = may, year =. Bayesian

    work page

  25. [25]

    Annual Review of Statistics and its Application , volume=

    Compositional data analysis , author=. Annual Review of Statistics and its Application , volume=. 2021 , publisher=

    work page 2021

  26. [26]

    Stochastic Environmental Research and Risk Assessment , author =

    Compositional time series analysis for. Stochastic Environmental Research and Risk Assessment , author =. 2018 , keywords =. doi:10.1007/s00477-018-1542-0 , abstract =

    work page doi:10.1007/s00477-018-1542-0 2018

  27. [27]

    Compositional

    Bacon-Shone, John and Grunsky, Eric , year =. Compositional

    work page

  28. [28]

    Aitchison's

    Greenacre, Michael and Grunsky, Eric and Bacon-Shone, John and Erb, Ionas and Quinn, Thomas , month = jan, year =. Aitchison's

    work page

  29. [29]

    Mathematical Geosciences , author =

    Geostatistics for. Mathematical Geosciences , author =. 2019 , keywords =. doi:10.1007/s11004-018-9769-3 , abstract =

    work page doi:10.1007/s11004-018-9769-3 2019

  30. [30]

    Journal of the Royal Statistical Society

    Review of. Journal of the Royal Statistical Society. Series A (General) , author =. 1986 , note =. doi:10.2307/2981571 , number =

    work page doi:10.2307/2981571 1986

  31. [31]

    Mathematical Geosciences , author =

    Compositional. Mathematical Geosciences , author =. 2020 , keywords =. doi:10.1007/s11004-020-09873-2 , abstract =

    work page doi:10.1007/s11004-020-09873-2 2020

  32. [32]

    Rasmussen, Carl Edward and Williams, Christopher K. I. , year =. Gaussian processes for machine learning , isbn =

    work page

  33. [33]

    , month = jan, year =

    MacQueen, J. , month = jan, year =. Some methods for classification and analysis of multivariate observations , volume =. Proceedings of the

    work page

  34. [34]

    Inferences from Multinomal Data: Learning about a bag of marbles

    Regression. Journal of the Royal Statistical Society Series B: Statistical Methodology , author =. 1996 , pages =. doi:10.1111/j.2517-6161.1996.tb02080.x , abstract =

    work page doi:10.1111/j.2517-6161.1996.tb02080.x 1996

  35. [35]

    Ester, Martin and Kriegel, Hans-Peter and Xu, Xiaowei , file =. A

    work page

  36. [36]

    , collaborator =

    McLachlan, Geoffrey J. , collaborator =. Mixture models: inference and applications to clustering , isbn =. 1988 , keywords =

    work page 1988

  37. [37]

    doi: 10.1080 /01621459.2013.794730

    Model. Journal of the American Statistical Association , author =. 2014 , pages =. doi:10.1080/01621459.2013.836975 , abstract =

    work page doi:10.1080/01621459.2013.836975 2014

  38. [38]

    Bayesian Analysis , author =

    Bayesian. Bayesian Analysis , author =. 2023 , file =. doi:10.1214/22-BA1320 , abstract =

    work page doi:10.1214/22-ba1320 2023

  39. [39]

    Statistical science : a review journal of the Institute of Mathematical Statistics , author =

    A. Statistical science : a review journal of the Institute of Mathematical Statistics , author =. 2012 , pmid =. doi:10.1214/12-STS392 , abstract =

    work page doi:10.1214/12-sts392 2012

  40. [40]

    Statistics and its interface , author =

    Penalized methods for bi-level variable selection , volume =. Statistics and its interface , author =. 2009 , pmid =

    work page 2009

  41. [41]

    2024 , url =

    GDP by State , howpublished =. 2024 , url =

    work page 2024

  42. [42]

    Mathematical Geology , author =

    Isometric. Mathematical Geology , author =. 2003 , file =

    work page 2003

  43. [43]

    Nearly unbiased variable selection under minimax concave penalty

    Zhang, Cun-Hui , month = feb, year =. Nearly unbiased variable selection under minimax concave penalty , url =. doi:10.48550/arXiv.1002.4734 , abstract =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1002.4734

  44. [44]

    Biometrika , author =

    Tuning parameter selectors for the smoothly clipped absolute deviation method , volume =. Biometrika , author =. 2007 , pmid =. doi:10.1093/biomet/asm053 , abstract =

    work page doi:10.1093/biomet/asm053 2007

  45. [45]

    Rand , title =

    Objective. Journal of the American Statistical Association , author =. 1971 , note =. doi:10.1080/01621459.1971.10482356 , abstract =

    work page doi:10.1080/01621459.1971.10482356 1971

  46. [46]

    Statistics in Medicine , author =

    Cluster detection of spatial regression coefficients , volume =. Statistics in Medicine , author =. 2017 , pages =. doi:10.1002/sim.7172 , abstract =

    work page doi:10.1002/sim.7172 2017

  47. [47]

    Bayesian Analysis , author =

    Bayesian. Bayesian Analysis , author =. 2016 , file =. doi:10.1214/14-BA925 , abstract =

    work page doi:10.1214/14-ba925 2016

  48. [48]

    Technometrics , author =

    Clustering. Technometrics , author =. 2012 , pages =. doi:10.1080/00401706.2012.657106 , abstract =

    work page doi:10.1080/00401706.2012.657106 2012

  49. [49]

    Statistica Neerlandica , author =

    Hierarchical clustering of spatially correlated functional data , volume =. Statistica Neerlandica , author =. 2012 , pages =. doi:10.1111/j.1467-9574.2012.00522.x , abstract =

    work page doi:10.1111/j.1467-9574.2012.00522.x 2012

  50. [50]

    Variable Selection via Nonconcave Penalized Likelihood and Its Oracle Properties , urldate =

    Jianqing Fan and Runze Li , journal =. Variable Selection via Nonconcave Penalized Likelihood and Its Oracle Properties , urldate =

    work page

  51. [51]

    Everitt, B. S. and Hand, D. J. , year =. Finite. doi:10.1007/978-94-009-5897-5 , keywords =

    work page doi:10.1007/978-94-009-5897-5

  52. [52]

    Inferences from Multinomal Data: Learning about a bag of marbles

    Discriminant. Journal of the Royal Statistical Society: Series B (Methodological) , author =. 1996 , note =. doi:10.1111/j.2517-6161.1996.tb02073.x , abstract =

    work page doi:10.1111/j.2517-6161.1996.tb02073.x 1996

  53. [53]

    Journal of Statistical Planning and Inference , author =

    Model-based classification using latent. Journal of Statistical Planning and Inference , author =. 2010 , keywords =. doi:10.1016/j.jspi.2009.11.006 , abstract =

    work page doi:10.1016/j.jspi.2009.11.006 2010

  54. [54]

    Biometrics , author =

    Model-. Biometrics , author =. 1993 , note =. doi:10.2307/2532201 , abstract =

    work page doi:10.2307/2532201 1993

  55. [55]

    doi: 10.1080 /01621459.2013.794730

    Latent. Journal of the American Statistical Association , author =. 2013 , pmid =. doi:10.1080/01621459.2013.789695 , abstract =

    work page doi:10.1080/01621459.2013.789695 2013

  56. [56]

    Journal of the American Statistical Association , author =

    Inference for. Journal of the American Statistical Association , author =. 2015 , note =

    work page 2015

  57. [57]

    Spectral Experts for Estimating Mixtures of Linear Regressions

    Chaganty, Arun Tejasvi and Liang, Percy , month = jun, year =. Spectral. doi:10.48550/arXiv.1306.3729 , abstract =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1306.3729

  58. [58]

    The State of the American Middle Class , year =

    work page

  59. [59]

    Journal of Statistical Software , year =

    Kurt Hornik , title =. Journal of Statistical Software , year =

    work page

  60. [60]

    Spatial Statistics , author =

    Transformed. Spatial Statistics , author =. 2015 , note =. doi:10.1016/j.spasta.2015.07.004 , abstract =

    work page doi:10.1016/j.spasta.2015.07.004 2015

  61. [61]

    Regularization and Variable Selection Via the Elastic Net

    Sparsity and. Journal of the Royal Statistical Society Series B: Statistical Methodology , author =. 2005 , pages =. doi:10.1111/j.1467-9868.2005.00490.x , abstract =

    work page doi:10.1111/j.1467-9868.2005.00490.x 2005

  62. [62]

    Biometrics , author =

    Simultaneous regression shrinkage, variable selection and clustering of predictors with. Biometrics , author =. 2008 , pmid =. doi:10.1111/j.1541-0420.2007.00843.x , abstract =

    work page doi:10.1111/j.1541-0420.2007.00843.x 2008

  63. [63]

    Journal of the American Statistical Association , author =

    Grouping pursuit through a regularization solution surface , volume =. Journal of the American Statistical Association , author =. 2010 , pmid =. doi:10.1198/jasa.2010.tm09380 , abstract =

    work page doi:10.1198/jasa.2010.tm09380 2010

  64. [64]

    Homogeneity in Regression

    Ke, Tracy and Fan, Jianqing and Wu, Yichao , month = mar, year =. Homogeneity in. doi:10.48550/arXiv.1303.7409 , abstract =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1303.7409

  65. [65]

    Biometrics , author =

    Pairwise variable selection for high-dimensional model-based clustering , volume =. Biometrics , author =. 2010 , pmid =. doi:10.1111/j.1541-0420.2009.01341.x , abstract =

    work page doi:10.1111/j.1541-0420.2009.01341.x 2010

  66. [66]

    and Lange, Kenneth , year=

    Splitting. Journal of Computational and Graphical Statistics , author =. 2015 , note =. doi:10.1080/10618600.2014.948181 , abstract =

    work page doi:10.1080/10618600.2014.948181 2015

  67. [67]

    Journal of the American Statistical Association , author =

    Variable. Journal of the American Statistical Association , author =. 2001 , note =

    work page 2001

  68. [68]

    W. R. Tobler , journal =. A Computer Movie Simulating Urban Growth in the Detroit Region , urldate =

    work page

  69. [69]

    Subramanian, S. V. and Kawachi, Ichiro , title =. Epidemiologic Reviews , volume =. 2004 , month =. doi:10.1093/epirev/mxh003 , url =

    work page doi:10.1093/epirev/mxh003 2004

  70. [70]

    Burchett and Simon Lewin and Ella R

    Helen E. Burchett and Simon Lewin and Ella R. Lavis and Lucy V. Mayhew and Atle Fretheim and Jonathan P. Oxman , title =. BMC Public Health , year =. doi:10.1186/1471-2458-13-1001 , url =

    work page doi:10.1186/1471-2458-13-1001

  71. [71]

    International Journal of Population Data Science , volume=

    Income inequalities in the risk of potentially avoidable hospitalisation for chronic obstructive pulmonary disease: a population data linkage analysis , author=. International Journal of Population Data Science , volume=. 2020 , publisher=

    work page 2020

  72. [72]

    , title =

    Snyder, John P. , title =. 1987 , publisher =

    work page 1987

  73. [73]

    and Hart, Peter E

    Duda, Richard O. and Hart, Peter E. and Stork, David G. , title =. 2001 , publisher =

    work page 2001

  74. [74]

    Biometrics , volume=

    Bayesian spatial homogeneity pursuit for survival data with an application to the SEER respiratory cancer data , author=. Biometrics , volume=. 2022 , publisher=

    work page 2022

  75. [75]

    Geographical Analysis , volume=

    Geographically weighted Cox regression for prostate cancer survival data in Louisiana , author=. Geographical Analysis , volume=. 2020 , publisher=

    work page 2020