The work is on detection of order restrictions in 2 × k cohort study. The existing tests in the literature are mostly based on subjective choice of column scores. Moreover, when the marginal distributions of the column totals differ, it is observed that the normal approximation to the null distribution (having no difference in the disease risks among the exposure categories) of the relevant test statistic for small samples is violated in most of the cases. As a remedy, utilizing such variation in column distributions, single contrast tests are proposed for this purpose. Performance of the proposed tests are assessed numerically and are compared with that of an existing competitor.
AgrestiACoullB. Order-restricted Tests for Stratified Comparisons of Binomial Proportions. Biometrics. 1996; 52(3): 1103–1111.
2.
HwangJPeddadaS. Confidence Interval Estimation Subject to Order Restrictions. Annal Stat. 1994; 22(1): 67–93.
3.
LiuQ. An Order-Directed Score Test for Trend in Ordered 2 × k Tables. Biometrics. 1998; 54(3): 1147–1154.
4.
PeddadaSPrescottKConawayM. Tests for Order Restrictions in Binary Response. Biometrics. 2001; 57(4): 1219–1227.
5.
SilvapulleM. On Tests against One-Sided Hypotheses in Some Generalized Linear Models. Biometrics. 1994; 50(3): 853–858.
6.
RobertsonTWrightFTDykstraRL. Order Restricted Statistical Inference. New York: Wiley; 1988.
7.
ArmitageP. Tests for Linear Trends in Proportions and Frequencies. Biometrics. 1955; 11(3): 375–86.
8.
CochranWG. Some Methods for Strengthening the Common χ 2test. Biometrics. 1954; 10(4): 417–451.
9.
MantelN. Chi-Square Tests with One Degree of Freedom; Extensions of the Mantel-Haenszel Procedure. J Amer Stat Assoc. 1963; 58(303): 690–700.
10.
GraubardBIKornEL. Choice of Column Scores for Testing Independence in Ordered 2 × k Contingency Tables. Biometrics. 1987; 43(2): 471–476.
11.
BandyopadhyayUChakrabartiP. Single Contrast Tests for Detecting Trends in Binomial Proportions. J Kor Stat Soc. 2013; 42(2): 235–246.
12.
HothornLAVaethMHothornT. Trend Tests for the Evaluation of Exposure-Response Relationships in Epidemiological Exposure Studies. Epidem Persp Innov. 2009; 6(1): 1–10.
13.
LeuraudKBenichouJ. A Comparison of Several Methods to Test for the Existence of a Monotonic Dose-Response Relationship in Clinical and Epidemiological Studies. Stat Med. 2001; 20(22): 3335–3351.
14.
LeuraudKBenichouJ. Tests for Monotonic Trend from Case-Control Data: Cochran-Armitage-Mantel Trend Test, Isotonic Regression and Single and Multiple Contrast Tests. Biomet. J. 2004; 46(6): 731–749.
15.
LeuraudKBenichouJ. A Comparison of Stratified and Adjusted Trend Tests for Binomial Proportions. Stat Med. 2006; 25(3): 529–535.
16.
GuptaPCMehtaHC. Cohort Study of all-Cause Mortality Among Tobacco users in Mumbai, India. BullWor Heal Organ2000; 78(7): 877–883.
