Using the bootstrap for statistical inference on random graphs

被引:10
|
作者
Thompson, Mary E. [1 ]
Ramirez, Lilia L. Ramirez [2 ,3 ]
Lyubchich, Vyacheslav [4 ]
Gel, Yulia R. [5 ]
机构
[1] Univ Waterloo, Waterloo, ON N2L 3G1, Canada
[2] ITAM, Mexico City, DF, Mexico
[3] Ctr Invest Matemat CIMAT, Mexico City, DF, Mexico
[4] Univ Maryland, Ctr Environm Sci, College Pk, MD USA
[5] Univ Texas Dallas, Richardson, TX 75083 USA
基金
加拿大自然科学与工程研究理事会; 美国国家科学基金会;
关键词
Bootstrap; Finite-sample inference; goodness-of-fit; networks; nonparametric methods; sampling; NETWORKS;
D O I
10.1002/cjs.11271
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
In this paper we propose a new nonparametric approach to network inference that may be viewed as a fusion of block sampling procedures for temporally and spatially dependent processes with the classical network methodology. We develop estimation and uncertainty quantification procedures for network mean degree using a patchwork sample and nonparametric bootstrap, under the assumption of unknown degree distribution. We provide a heuristic justification of asymptotic properties of the proposed patchwork sampling and present cross-validation methodology for selecting an optimal patch size. We validate the new patchwork bootstrap on simulated networks with short- and long-tailed mean degree distributions, and revisit the Erdos collaboration data to illustrate the proposed methodology. The Canadian Journal of Statistics 44: 3-24; 2016 (c) 2015 Statistical Society of Canada Resume Les auteurs proposent une approche non parametrique pour l'inference dans les reseaux qui peut etre decrite comme une fusion entre la methodologie classique pour les reseaux et des procedures d'echantillonnage par bloc pour des processus presentant une dependance spatiale et temporelle. Ils developpent des procedures permettant d'estimer le degre moyen du reseau et d'en mesurer l'incertitude grace a l'echantillonnage en mosai que et au bootstrap non parametrique, sous l'hypothese que la distribution du degre est inconnue. Les auteurs presentent une justification heuristique des proprietes asymptotiques de l'echantillonnage en mosai que propose et decrivent une methodologie de validation croisee afin de choisir la taille optimale des pieces de la mosai que. Ils valident le bootstrap pour la mosai que en simulant des reseaux dont la distribution du degre moyen presente des queues lourdes ou legeres. Ils illustrent egalement leur methodologie a l'aide des donnees de collaboration d'Erdos. La revue canadienne de statistique 44: 3-24; 2016 (c) 2015 Societe statistique du Canada
引用
收藏
页码:3 / 24
页数:22
相关论文
共 50 条
  • [1] Statistical Inference on Random Dot Product Graphs: a Survey
    Athreya, Avanti
    Fishkind, Donniell E.
    Minh Tang
    Priebe, Carey E.
    Park, Youngser
    Vogelstein, Joshua T.
    Levin, Keith
    Lyzinski, Vince
    Qin, Yichen
    Sussman, Daniel L.
    [J]. JOURNAL OF MACHINE LEARNING RESEARCH, 2018, 18
  • [2] Statistical inference on random dot product graphs: A survey
    Athreya, Avanti
    Fishkind, Donniell E
    Tang, Minh
    Priebe, Carey E
    Park, Youngser
    Vogelstein, Joshua T.
    Levin, Keith
    Lyzinski, Vince
    Qin, Yichen
    Sussman, Daniel L
    [J]. Journal of Machine Learning Research, 2018, 18 : 1 - 92
  • [3] USING THE SMOOTHED BOOTSTRAP FOR STATISTICAL INFERENCE FOR MARKOV CHAINS
    Polansky, Alan M.
    [J]. PAKISTAN JOURNAL OF STATISTICS, 2009, 25 (04): : 553 - 570
  • [4] Simple bootstrap statistical inference using the SAS system
    Cole, SR
    [J]. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE, 1999, 60 (01) : 79 - 82
  • [5] STATISTICAL INFERENCE WITH PLSC USING BOOTSTRAP CONFIDENCE INTERVALS
    Aguirre-Urreta, Miguel I.
    Ronkko, Mikko
    [J]. MIS QUARTERLY, 2018, 42 (03) : 1001 - +
  • [6] RANDOM ATTRIBUTED GRAPHS FOR STATISTICAL INFERENCE FROM CONTENT AND CONTEXT
    Gorin, A. L.
    Priebe, C. E.
    Grothendieck, J.
    [J]. 2010 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, 2010, : 5430 - 5433
  • [7] Statistical inference on graphs
    Biau, Gerard
    Bleakley, Kevin
    [J]. STATISTICS & RISK MODELING, 2006, 24 (02) : 209 - 232
  • [8] Statistical inference on attributed random graphs: Fusion of graph features and content
    Grothendieck, John
    Priebe, Carey E.
    Gorin, Allen L.
    [J]. COMPUTATIONAL STATISTICS & DATA ANALYSIS, 2010, 54 (07) : 1777 - 1790
  • [9] Bootstrap percolation in inhomogeneous random graphs
    Amini, Hamed
    Fountoulakis, Nikolaos
    Panagiotou, Konstantinos
    [J]. ADVANCES IN APPLIED PROBABILITY, 2024, 56 (01) : 156 - 204
  • [10] Bootstrap percolation in random geometric graphs
    Falgas-Ravry, Victor
    Sarkar, Amites
    [J]. ADVANCES IN APPLIED PROBABILITY, 2023, 55 (04) : 1254 - 1300