Abstract：Complex network is the skeleton of the complex system. It composes of nodes and edges. Most of the networks have some important properties such as the power law degree distribution and the small world effect. Usually, it is not easy to justify the scale free degree distribution and estimate the parameters of the scale free distribution and to quantify both global and local network structure at the same time. In this paper, we will review the main statistics inference methods of complex networks, such as the estimation of parameters in power law distribution and exponential random graph model. Moreover, we also comment these methods from a statistical mathematical standpoint.
杨宝莹, 胡延庆. 统计推断方法在复杂网络中的应用[J]. 复杂系统与复杂性科学, 2014, 11(1): 67-76.
YANG Bao-ying, HU Yan-qing. The Application of Statistical Inference in Complex Networks[J]. Complex Systems and Complexity Science, 2014, 11(1): 67-76.
[1] Watts D J, Strogatz S H. Collective dynamics of small-world networks[J]. Nature, 1998, 393 （6684):409-410. [2] Barabási A L, Albert R. Emergence of Scaling in Random Networks[J]. Science, 1999, 286（15):509-512. [3] Clauset A, Shalizi C R, Newman M E J. Power law distribution in empirical data[J]. SIAM Review, 2009, 51（4):661-703. [4] Muniruzzaman A N M. On measures of location and dispersion and test of hypothesis on a Pareto distribution[J]. Bulletin of the Calcutta Statistical Association, 1957,7, 115-123. [5] Stoev S A, Michailidis G, Taqqu M S. Estimating heavy-tail exponents through max self-similarity[J]. IEEE Trans Inf Theory, 2011, 57（3):1636-1651. [6] Efron B, Tibshirani R J. An Introduction to the Bootstrap[M]. New York:Chapman and Hall,1993. [7] 谢益辉,朱钰.Bootstrap方法的历史发展和前沿研究[J].统计与信息论坛,2008,23（2):90-96 Xie yihui, Zhu Yu. The development and frontier research of the Bootstrap method[J]. Statistics and Information Forum, 2008, 23（2):90-96. [8] Vuong Q H. Likelihood ratio tests for model selection and non-nested hypotheses[J]. Econometrica, 1989, 57:307-333. [9] Robins G, Pattison P, Kalish Y, et al. An introduction to exponential random graph （p*) models for social metworks[J]. Social Networks, 2007,29:173-191. [10] Rapoport A. Spread of information through a population with a socio-structural bias（I). assumption of transitivity[J]. Bulletin of Mathematical Biophysics, 1953, 15:523-533. [11] Erdos P, Renyi A. On random graphs I[J]. Publicationes Mathematicae （Debrecen), 1959, 6:290-297. [12] Holland P W, Leinhardt S. An exponential family of probability distributions for directed graphs[J]. Journal of the American Statistical Association, 1981, 76 （373):33-50. [13] Van Duijn M A J, Snijders T A B, Zijlstra B H. P2:a random effects model with covariates for directed graphs[J]. Statistica Neerlandica, 2004, 58:234 254. [14] Frank O, Strauss D. Markov graphs[J]. Journal of the American Statistical Association, 1986, 81 （395):832-842. [15] Snijders T A B, Pattison P, Robins G L,et al. New specifications for exponential random graph models[J]. Sociological M ethodology, 2006, 36:99-153. [16] Anderson C J, Wasserman S, Grouch B. A P*primer:logit models for social network[J].Social Networks,1999,21:37-66. [17] Strauss D, Ikeda M. Pseudo-likelihood estimation for social networks[J]. Journal of the American Statistical Association, 1990,85:204-212. [18] Geyer C J, Thompson E A.Constrained Monte Carlo maximum likelihood for dependent data[J].Journal of the Royal Statistical Society, Series B, 1992, 54:657-699 [19] Snijders T A B. Markov chain Monte Carlo estimation of exponential random graph models[DB/OL].[2013-07-01].http://www.cmu.edu/joss/content/articles/volume3/snijders.pdf. [20] Handcock M S, Hunter D R, Butts C T, et al. Software tools for the representation, visualization, analysis and simulation of network data[J]. Journal of Statistical Software, 2008, 24（1):1548.
