Statistical Inference for Sex Ratio at Birth

Xiaochun Qiao, Peking University
Chirayath Suchindran, University of North Carolina at Chapel Hill

This paper is to explore how to test if SRB is abnormal, less than the 103 or greater than 107, the cut-point for abnormal SRB. In this paper, we, first, give the formula for calculating the standard error of SRB based on SRB. Second, the standard errors under different SRB and different sample size are calculated, and the confidence intervals under 95% confidence are estimated. Third, we show the one sample t test for the difference between SRB and a given value, usually 105, 103, or 107. Fourth, we discuss about the two sample t test for the difference between two SRBs, by using one of the two chi-square test methods. The last, we will discuss the determination of the sample size for testing if the SRB is normal (103-107) or if the SRB is abnormal (mostly greater than 107).