Statistical inference under order restrictions on both rows and columns of a matrix, with an application in toxicology
Eric Teoh, Abraham Nyska, Uri Wormser, Shyamal D. Peddada
Abstract
We present a general methodology for performing statistical inference on the components of a real-valued matrix parameter for which rows and columns are subject to order restrictions. The proposed estimation procedure is based on an iterative algorithm developed by Dykstra and Robertson (1982) for simple order restriction on rows and columns of a matrix. For any order restrictions on rows and columns of a matrix, sufficient conditions are derived for the algorithm to converge in a single application of row and column operations. The new algorithm is applicable to a broad collection of order restrictions. In practice, it is easy to design a study such that the sufficient conditions derived in this paper are satisfied. For instance, the sufficient conditions are satisfied in a balanced design. Using the estimation procedure developed in this article, a bootstrap test for order restrictions on rows and columns of a matrix is proposed. Computer simulations for ordinal data were performed to compare the proposed test with some existing test procedures in terms of size and power. The new methodology is illustrated by applying it to a set of ordinal data obtained from a toxicological study.
First Page:
Show
Hide
Primary Subjects: 62F10
Secondary Subjects: 62G09, 62G10
Keywords: linked parameters; matrix partial order; maximally-linked subgraph; order-restriction; ordinal data; simple order; simple tree order; umbrella order
Full-text: Open access
Links and Identifiers
Permanent link to this document: http://projecteuclid.org/euclid.imsc/1207058264
Digital Object Identifier: doi:10.1214/193940307000000059
References
[1] Agresti, A. and Coull, B. A. (2002). The analysis of contingency tables under inequality constraints. J. Statist. Plann. Inference 107 45–73.
Mathematical Reviews (MathSciNet):
MR1927754
Zentralblatt MATH:
1016.62075
Digital Object Identifier: doi:10.1016/S0378-3758(02)00243-4
[2] Barlow, R., Bartholomew, D., Bremmer, J. and Brunk, H. (1972). Statistical Inference Under Order Restrictions, 3rd ed. ed. Wiley, New York.
[3] Berger, V., Permutt, T. and Ivanova, A. (1998). Convex hull test for ordered categorical data. Biometrics 54 1541–1550.
[4] Cohen, A., Kemperman, J. and Sackrowitz, H. (2000). Properties of likelihood inference for order restricted models. J. Multivariate Anal. 72 50–77.
Mathematical Reviews (MathSciNet):
MR1747423
Zentralblatt MATH:
0978.62046
Digital Object Identifier: doi:10.1006/jmva.1999.1847
[5] Cohn, S., Schloemann, S., Tessner, T., Seibert, K. and Stenson, W. (1997). Crypt stem cell survival in the mouse intestinal epithelium is regulated by prostaglandins synthesized through cyclooxygenase-1. J. Clinical Investigation 99 1367–1379.
[6] Conaway, M. R., Dunbar, S. and Peddada, S. D. (2004). Designs for single- or multiple-agent phase I trials. Biometrics 60 661–669.
Mathematical Reviews (MathSciNet):
MR2089441
Digital Object Identifier: doi:10.1111/j.0006-341X.2004.00215.x
JSTOR: links.jstor.org
[7] Dykstra, R. L. and Robertson, T. (1982). An algorithm for isotonic regression for two or more independent variables. Ann. Statist. 10 708–716.
Mathematical Reviews (MathSciNet):
MR663427
Zentralblatt MATH:
0485.65099
Digital Object Identifier: doi:10.1214/aos/1176345866
Project Euclid: euclid.aos/1176345866
[8] Franck, W. E. (1984). A likelihood ratio test for stochastic ordering. J. Amer. Statist. Assoc. 79 686–691.
Mathematical Reviews (MathSciNet):
MR763587
Zentralblatt MATH:
0553.62043
Digital Object Identifier: doi:10.2307/2288417
JSTOR: links.jstor.org
[9] Grove, D. M. (1980). A test of independence against a class of ordered alternatives in a 2×C contingency table. J. Amer. Statist. Assoc. 75 454–459.
[10] He, L., Liu, L., Hahn, E. and Gamelli, R. (2001). he expression of cyclooxygenase and the production of prostaglandin E2 in neutrophils after burn injury and infection. J. Burn Care Rehabilitation 22 58–64.
[11] Hwang, J. T. G. and Peddada, S. D. (1994). Confidence interval estimation subject to order restrictions. Ann. Statist. 22 67–93.
Mathematical Reviews (MathSciNet):
MR1272076
Zentralblatt MATH:
0796.62033
Digital Object Identifier: doi:10.1214/aos/1176325358
Project Euclid: euclid.aos/1176325358
[12] Lee, C.-I. C. (1988). Quadratic loss of order restricted estimators for treatment means with a control. Ann. Statist. 16 751–758.
Mathematical Reviews (MathSciNet):
MR947575
Digital Object Identifier: doi:10.1214/aos/1176350833
Project Euclid: euclid.aos/1176350833
[13] Lee, J., Mukhtar, H., Bickers, D., Kopelovich, L. and Athar, M. (2003). Cyclooxygenases in the skin: Pharmacological and toxicological implications. Toxicology and Applied Pharmacology 192 294–306.
[14] Lin, H., Lin, T., Cheung, W., Nian, G., Tseng, P., Chen, S., Chen, J., Shyue, S., Liou, J., Wu, C. and Wu, K. (2002). Cyclooxygenase-1 and bicistronic cyclooxygenase-1/prostacyclin synthase gene transfer protect against ischemic cerebral infarction. Circulation 105 1962–1969.
[15] Lehmann, E. L. (1983). Theory of Point Estimation. Wiley, New York.
Mathematical Reviews (MathSciNet):
MR702834
[16] Nair, V. N. (1987). Chi-squared-type tests for ordered alternatives in contingency tables. J. Amer. Statist. Assoc. 82 283–291.
Mathematical Reviews (MathSciNet):
MR883357
Zentralblatt MATH:
0688.62033
Digital Object Identifier: doi:10.2307/2289165
JSTOR: links.jstor.org
[17] Peddada, S. D., Prescott, K. E. and Conaway, M. (2001). Tests for order restrictions in binary data. Biometrics 57 1219–1227.
Mathematical Reviews (MathSciNet):
MR1973822
Digital Object Identifier: doi:10.1111/j.0006-341X.2001.01219.x
JSTOR: links.jstor.org
[18] Peddada, S. D., Dunson, D. B. and Tan, X. (2005). Estimation of order-restricted means from correlated data. Biometrika 92 703–715.
Mathematical Reviews (MathSciNet):
MR2202656
Zentralblatt MATH:
1152.62348
Digital Object Identifier: doi:10.1093/biomet/92.3.703
[19] Præstgaard, J. T. and Huang, J. (1996). Asymptotic theory for nonparametric estimation of survival curves under order restrictions. Ann. Statist. 24 1679–1716.
[20] Robertson, T. and Wright, F. T. (1981). Likelihood ratio tests for and against a stochastic ordering between multinomial populations. Ann. Statist. 9 1248–1257.
Mathematical Reviews (MathSciNet):
MR630107
Digital Object Identifier: doi:10.1214/aos/1176345641
Project Euclid: euclid.aos/1176345641
[21] Robertson, T., Wright, F. T. and Dykstra, R. L. (1988). Order Restricted Statistical Inference. Wiley, Chichester.
Mathematical Reviews (MathSciNet):
MR961262
[22] Silvapulle, M. J. (1997). A curious example involving the likelihood ratio test against one-sided hypotheses. American Statistician 51 178–180.
[23] Silvapulle, M. J. and Sen, P. K. (2005). Constrained Statistical Inference. Wiley, Hoboken, NJ.
Mathematical Reviews (MathSciNet):
MR2099529
Zentralblatt MATH:
1077.62019
[24] Vane, J., Bakhle, Y. and Botting, R. (1998). Cyclooxygenases 1 and 2. Annual Review of Pharmacology and Toxicology 38 97–120.
[25] Wang, Y. (1996). A likelihood ratio test against stochastic ordering in several populations. J. Amer. Statist. Assoc. 91 1676–1683.
Mathematical Reviews (MathSciNet):
MR1439109
Zentralblatt MATH:
0881.62019
Digital Object Identifier: doi:10.2307/2291595
JSTOR: links.jstor.org
[26] Wormser, U., Langenbach, R., Peddada, S., Sintov, A., Brodsky, B. and Nyska, A. (2004). Reduced sulfur mustard-induced skin toxicity in cyclooxygenase-2 knockout and celecoxib-treated mice. Toxicology and Applied Pharmacology 200 40–47.
Institute of Mathematical Statistics Collections
