Sufficiency and Necessity for the Identifiability of An Important Type of Ancillary Information in A Causal Model

Acta Scientiarum Naturalium Universitatis Pekinensis

Sufficiency and Necessity for the Identifiability of An Important Type of Ancillary Information in A Causal Model

TONG Xingwei¹, ZHENG Zhongguo¹, ZHANG Yanyan², CUI Xianwei¹

¹School of Mathematical Sciences, Peking University, Beijing, 100871; ²Tianjin Normal University, Tianjin, 301700

Received:2001-06-30 Online:2002-09-20 Published:2002-09-20

一类重要辅助信息在一类因果模型中可识别的充要条件

童行伟¹, 郑忠国¹, 张艳艳², 崔现伟¹

¹北京大学数学科学学院，北京，100871；²天津师范大学，天津，301700

Abstract

Abstract: Much of epidemiological and social science research is devoted to estimation of causal effect. Generally, the causal effect is not identifiable without any constraint which is called the ancillary information in this paper. First the definition of the identifiability for the ancillary information is proposed and then sufficient and necessary conditions for the identifiability of an important type of ancillary information are discussed. Finally, the results are expanded to a more common case.

Key words: identifiability, causal effect, ancillary information

摘要： 在流行病学和社会科学中，大量的研究是估计因果效应。一般而言，没有任何假设的条件下，因果效应无法获得，即不给因果变量任何辅助信息，因果效应无法识别。作者提出了辅助信息可识别的定义，并且针对一类比较普遍且重要的辅助信息得出它在一类因果模型中对4种因果效应可识别的充要条件，最后将这结果推广到一般情况。

关键词: 可识别, 因果效应, 辅助信息

CLC Number:

O213
R18

TONG Xingwei,ZHENG Zhongguo,ZHANG Yanyan,CUI Xianwei. Sufficiency and Necessity for the Identifiability of An Important Type of Ancillary Information in A Causal Model[J]. Acta Scientiarum Naturalium Universitatis Pekinensis.

童行伟, 郑忠国, 张艳艳, 崔现伟. 一类重要辅助信息在一类因果模型中可识别的充要条件[J]. 北京大学学报（自然科学版）.

[1]	CHEN Yanguang. An Approach Based on Euler’s Formula to Estimating Gini Coefficient of Pareto Distribution [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2018, 54(6): 1283-1289.
[2]	HAN Fang,WU Jingchen,XU Jiangfeng,DENG Minghua. Searching for Differentially Expressed Genes by PLS-VIP Method [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2009, 45(1): 1-5.
[3]	HE Shuyuan,ZHAO Yu,FANG Xiangzhong. Equipment Reliability Estimation under Different Experiment Periods and Environments [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2008, 44(3): 331-334.
[4]	ZHAO Hui,ZHENG Zhongguo,XU Jing. Causal Inference in the Models with Hidden Variables and Selection Bias [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2006, 42(5): 584-589.
[5]	CHEN Jiading,LI Ji. On the Lower Confidence Limits for the Structure Reliability [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2004, 40(5): 702-711.
[6]	TANG Hongmin,SUN Zhaoli. A Cointegration Analysis among Four Exchange Rates and Four Stock Indices [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2003, 39(5): 601-606.
[7]	WANG Yanhua,HE Shuyuan,MEI Changlin. Fisher Information for Censored Data from Location Scale Families [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2003, 39(5): 612-619.
[8]	XU Jing,ZHENG Zhongguo. The Structure and Determination of the Set of Reversible Edges in a Directed Acyclic Graph [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2003, 39(1): 6-13.
[9]	LI Xiande,YANG Jingping. Compound Poisson Approximation of Individual Risk Models [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2001, 37(5): 609-617.
[10]	YANG Baohui,SUN Shanze. Regression Estimation Using Cluster Correlation [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 1999, 35(1): 18-22.
[11]	ZHANG Xiao. Confidence Limits about The Mean Time between the Failures (MTBF) in Assurance Test [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 1999, 35(1): 23-30.
[12]	ZHENG Zhongguo,JIN Hua. Lower Confidence Limits in Censored Case [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 1995, 31(1): 27-39.