Exact and approximate unconditional confidence intervals for proportion difference in the presence of incomplete data

Confidence interval (CI) construction with respect to proportion/rate difference for paired binary data has become a standard procedure in many clinical trials and medical studies. When the sample size is small and incomplete data are present, asymptotic CIs may be dubious and exact CIs are not yet available. In this article, we propose exact and approximate unconditional test-based methods for constructing CI for proportion/rate difference in the presence of incomplete paired binary data. Approaches based on one- and two-sided Wald's tests will be considered. Unlike asymptotic CI estimators, exact unconditional CI estimators always guarantee their coverage probabilities at or above the pre-specified confidence level. Our empirical studies further show that (i) approximate unconditional CI estimators usually yield shorter expected confidence width (ECW) with their coverage probabilities being well controlled around the pre-specified confidence level; and (ii) the ECWs of the unconditional two-sided-test-based CI estimators are generally narrower than those of the unconditional one-sided-test-based CI estimators. Moreover, ECWs of asymptotic CIs may not necessarily be narrower than those of two-sided-based exact unconditional CIs. Two real examples will be used to illustrate our methodologies.

[1]  Hale Je,et al.  Identification of ampoules. , 1973 .

[2]  S C Choi,et al.  Comparing incomplete paired binomial data under non-random mechanisms. , 1988, Statistics in medicine.

[3]  Stuart R. Lipsitz,et al.  A Model for Binary Time Series Data with Serial Odds Ratio Patterns , 1995 .

[4]  Samy Suissa,et al.  Exact unconditional sample sizes for the 2×2 binomial trial , 1985 .

[5]  B. Foxman,et al.  Response from Drs. Foxman and Frerichs , 1986 .

[6]  Unconditional Exact Tests for Equivalence or Noninferiority for Paired Binary Endpoints , 2001, Biometrics.

[7]  Samy Suissa,et al.  Exact Unconditional Sample Sizes for the 2 Times 2 Binomial Trial , 1985 .

[8]  Niansheng Tang,et al.  Exact Tests for Comparing Two Paired Proportions with Incomplete Data , 2004 .

[9]  Devan V Mehrotra,et al.  A Cautionary Note on Exact Unconditional Inference for a Difference between Two Independent Binomial Proportions , 2003, Biometrics.

[10]  I S Chan,et al.  Test‐Based Exact Confidence Intervals for the Difference of Two Binomial Proportions , 1999, Biometrics.

[11]  S. C. Choi,et al.  Practical Tests for Comparing Two Proportions with Incomplete Data , 1982 .

[12]  Man-Lai Tang,et al.  Sample Size Determination for Establishing Equivalence/Noninferiority via Ratio of Two Proportions in Matched–Pair Design , 2002, Biometrics.

[13]  Xun Chen,et al.  A quasi‐exact method for the confidence intervals of the difference of two independent binomial proportions in small sample cases , 2002, Statistics in medicine.

[14]  J. Fleiss,et al.  Confidence intervals vs significance tests: quantitative interpretation. , 1986, American journal of public health.

[15]  Q. Mcnemar.  Note on the sampling error of the difference between correlated proportions or percentages , 1947, Psychometrika.

[16]  R G Newcombe,et al.  Improved confidence intervals for the difference between binomial proportions based on paired data. , 1998, Statistics in medicine.

[17]  Debabrata Basu,et al.  On the Elimination of Nuisance Parameters , 1977 .

[18]  R. Newcombe,et al.  Interval estimation for the difference between independent proportions: comparison of eleven methods. , 1998, Statistics in medicine.

[19]  R. Newcombe.  Two-sided confidence intervals for the single proportion: comparison of seven methods. , 1998, Statistics in medicine.

[20]  Man-Lai Tang,et al.  On improved EM algorithm and confidence interval construction for incomplete r , 2007, Comput. Stat. Data Anal..

[21]  M. Gardner,et al.  Confidence intervals rather than P values: estimation rather than hypothesis testing. , 1986, British medical journal.