
网络结构数据空间回归模型的平均估计
Model Averaging Estimation of Spatial Regression Model with Network Structure Data
采用空间误差模型对数据的网络结构关系进 行刻画, 考虑了空间误差模型的~S-AIC 和~S-BIC 模型平均估计方法, 证明了~S-AIC 和~S-BIC 估计的相合性和渐近正态性. 通过蒙特卡洛模拟试验, 研究了所提估计的有限样本性质, 模拟结果显示,~S-AIC 和~S-BIC 模型平均估计表现优于~AIC 和~BIC 模型选择估计. 利用文章所提方法, 对~QQ 用户数据进行实证分析, 说明了所提方法在实际问题中的应用价值.
In this paper, we adopt the spatial error model to depict the network structure relationship between individuals, we consider S-AIC and S-BIC model averaging estimation of the spatial error model and show the consistency and asymptotical normality of the S-AIC and S-BIC estimators. We conduct Monte Carlo experiments to investigate the finite sample properties of the proposed estimators. The simulation results show that the S-AIC and S-BIC model averaging estimators perform better than the AIC and BIC model selection estimators. We analyze the QQ user data set to illustrate the application of the proposed method.
/
〈 |
|
〉 |