McCullagh & Nelder, Generalised Linear Models, 2nd edition. Chapman & Hall, 1989.

本书概括了关于线性模型的假设，检验以及怎么用以有的数据选择feature建立线性模型。线性回归，英文原版

This is the textbook used in the course STAT347, in Department of Statistics, the University of Chicago.

REGHDFE：具有多个固定效应的线性回归 当前版本： 5.6.8 03mar2019 （当前SSC版本： 5.6.8 03mar2019 ） 跳转到： 最近更新 版本5.7.3 13nov2019 ： 使用压缩选项（＃194）修复了罕见错误。 版本也已提交给SSC。 版本5.7.0 20mar2019 ： 用户不再需要运行reghdfe, compile安装后即可进行reghdfe, compile 。 如果出现错误“类FixedEffects未定义”，请升级到该版本或运行reghdfe, compile 版本5.6.8 03mar2019 ： 发布软件包，用于具有固定效果的Poisson模型。 如果您使用左侧的log(y)运行回归，请使用此选项。 reghdfe的稳定版本，也在SSC上。 版本5.6.2 10feb2019 ： 运行所需的最低要求版本 版本5.6

破产机器学习 破产数据研究的目的是为给定数据确定预测破产的最佳分类方法。 破产数据是从COMPUSTAT收集的1980年至2000年的数据，其中有5436个观察值和13个变量。 9个基于会计的变量和1个市场变量是：R1：WC / TA，营运资金/总资产R2：RE / TA，未分配利润/总资产R3：EBIT / TA，息税前利润/总资产R4：ME / TL，权益/总负债的市场价值R5：S / TA，销售/总资产R6：TL / TA，总负债/总资产R7：CA / CL，流动资产/流动负债R8：NI / TA，净收入/总资产R9：破产成本，对数（销售）R10：市值，对数（绝对（价格）*流通股数/ 1000） 对于本研究，由于没有明显的破产趋势，因此可以假定可以将多年来的数据汇总在一起并进行研究。 在这13个变量中，其中一个是“ DLRSN”-一种表示默认值的分类变量，即预测的因变量。 总体而

Applied Regression Analysis and Generalized Linear Models,3rd Applied Regression Analysis and Generalized Linear Models,3rd

Generalized Linear Models，比另一个19MB的资源少了很多影印的黑墨

Generalized Linear Models书本，希望有帮助

The textbook for STAT343 in Dept. of Statistics in the University of Chicago.

The original purpose of the book was to present a unified theoretical and conceptual framework for statistical modelling in a way that was accessible to undergraduate students and researchers in other fields. The second edition was expanded to include nominal and ordinal logistic regression, survival analysis and analysis of longitudinal and clustered data. It relied more on numerical methods, visualizing numerical optimization and graphical methods for exploratory data analysis and checking model fit. The third edition added three chapters on Bayesian analysis for general- ized linear models. To help with the practical application of generalized linear models, Stata, R and WinBUGS code were added. This fourth edition includes new sections on the common problems of model selection and non-linear associations. Non-linear associations have a long history in statistics as the first application of the least squares method was when Gauss correctly predicted the non-linear orbit of an asteroid in 1801. Statistical methods are essential for many fields of research, but a widespread lack of knowledge of their correct application is creating inaccu- rate results. Untrustworthy results undermine the scientific process of using data to make inferences and inform decisions. There are established practices for creating reproducible results which are covered in a new Postface to this edition.