Estimation following selection of the largest of two normal means

This paper considers the problem of estimation when one of a number of populations, assumed normal with known common variance, is selected on the basis of it having the largest observed mean. Conditional on selection of the population, the observed mean is a biased estimate of the true mean. This problem arises in the analysis of clinical trials in which selection is made between a number of experimental treatments that are compared with each other either with or without an additional control treatment. Attempts to obtain approximately unbiased estimates in this setting have been proposed by Shen [2001. An improved method of evaluating drug effect in a multiple dose clinical trial. Statist. Medicine 20, 1913–1929] and Stallard and Todd [2005. Point estimates and confidence regions for sequential trials involving selection. J. Statist. Plann. Inference 135, 402–419]. This paper explores the problem in the simple setting in which two experimental treatments are compared in a single analysis. It is shown that in this case the estimate of Stallard and Todd is the maximum-likelihood estimate (m.l.e.), and this is compared with the estimate proposed by Shen. In particular, it is shown that the m.l.e. has infinite expectation whatever the true value of the mean being estimated. We show that there is no conditionally unbiased estimator, and propose a new family of approximately conditionally unbiased estimators, comparing these with the estimators suggested by Shen.     

Maximum-likelihood estimation and presentation for the interaction between treatments in observational studies with a dichotomous outcome

In observational studies for the interaction between treatments, one needs to estimate and present both the treatment effects and the interaction to learn the significance of the interaction to the treatment effects. In this article, we estimate the treatment effects and the interaction jointly by using only one logistic model and based on maximum-likelihood. We present the interaction by (1) point estimate and confidence interval of the interaction, (2) point estimate and confidence region of (treatment effect, interaction), and (3) point estimate and confidence interval of the interaction when the maximum-likelihood estimate of one treatment effect falls into specified range.     

Reducing Bias and Mean Squared Error Associated With Regression-Based Odds Ratio Estimators

Ratio estimators of effect are ordinarily obtained by exponentiating maximum-likelihood estimators (MLEs) of log-linear or logistic regression coefficients. These estimators can display marked positive finite-sample bias, however. We propose a simple correction that removes a substantial portion of the bias due to exponentiation. By combining this correction with bias correction on the log scale, we demonstrate that one achieves complete removal of second-order bias in odds ratio estimators in important special cases. We show how this approach extends to address bias in odds or risk ratio estimators in many common regression settings. We also propose a class of estimators that provide reduced mean bias and squared error, while allowing the investigator to control the risk of underestimating the true ratio parameter. We present simulation studies in which the proposed estimators are shown to exhibit considerable reduction in bias, variance, and mean squared error compared to MLEs. Bootstrapping provides further improvement, including narrower confidence intervals without sacrificing coverage.     

One-sided confidence intervals in discrete distributions

One-sided confidence intervals in the binomial, negative binomial, and Poisson distributions are considered. It is shown that the standard Wald interval suffers from a serious systematic bias in the coverage and so does the one-sided score interval. Alternative confidence intervals with better performance are considered. The coverage and length properties of the confidence intervals are compared through numerical and analytical calculations. Implications to hypothesis testing are also discussed.     

Simultaneous inference for treatment regimes

A viable dynamic treatment regime refers to decisions regarding how different treatments and dose levels are tailored through time to match with the patient’s health status. In the therapy for cancer or diseases that require multiple stages of treatments, the effect of preceding treatment (such as growing back of solid tumors or regime-related toxicity) critically influences the selection of treatment in the following stage. So far, analyses of dynamic regimes mainly focus on marginal mean models under the assumption of sequential randomization in a clinical trial. Inference conclusions regarding multiple regimes are normally conservative due to different combinations in the formation of treatment regimes. In this article, we propose a simultaneous confidence interval method to identify treatment regimes that are significantly different from the bulk of the treatment combinations. The new method is applied to analyze the dynamic treatment regimes (DTRs) prostate cancer trial which includes treatment combinations of four chemotherapies at multiple stages. The new method detects a discernible effect of the regime taxane–estramustine–carboplatin (TEC) followed by cyclophosphamide, vincristine, and dexamethasone (CVD), when it is compared with other four regimes starting with CVD.     

Smooth tests of fit for censored gmma samples

In this paper properties of two estimators of Cpm are investigated in terms of changes in the process mean and variance. The bias and mean squared error of these estimators are derived. It can be shown that the estimate of Cpm proposed by Chan, Cheng and Spiring (1988) has smaller bias than the one proposed by Boyles (1991) and also has a smaller mean squared error under certain conditions. Various approximate confidence intervals for Cpm are obtained and are compared in terms of coverage probabilities, missed rate and average interval width.     

Nonparametric confidence intervals for Tmax in sequence-stratified crossover studies

Tmax is the time associated with the maximum serum or plasma drug concentration achieved following a dose. While Tmax is continuous in theory, it is usually discrete in practice because it is equated to a nominal sampling time in the noncompartmental pharmacokinetics approach. For a 2-treatment crossover design, a Hodges-Lehmann method exists for a confidence interval on treatment differences. For appropriately designed crossover studies with more than two treatments, a new median-scaling method is proposed to obtain estimates and confidence intervals for treatment effects. A simulation study was done comparing this new method with two previously described rank-based nonparametric methods, a stratified ranks method and a signed ranks method due to Ohrvik. The Normal theory, a nonparametric confidence interval approach without adjustment for periods, and a nonparametric bootstrap method were also compared. Results show that less dense sampling and period effects cause increases in confidence interval length. The Normal theory method can be liberal (i.e. less than nominal coverage) if there is a true treatment effect. The nonparametric methods tend to be conservative with regard to coverage probability and among them the median-scaling method is least conservative and has shortest confidence intervals. The stratified ranks method was the most conservative and had very long confidence intervals. The bootstrap method was generally less conservative than the median-scaling method, but it tended to have longer confidence intervals. Overall, the median-scaling method had the best combination of coverage and confidence interval length. All methods performed adequately with respect to bias.     

Robust confidence regions for multinomial probabilities

Generally, confidence regions for the probabilities of a multinomial population are constructed based on the Pearson χ² statistic. Morales et al. (Bootstrap confidence regions in multinomial sampling. Appl Math Comput. 2004;155:295–315) considered the bootstrap and asymptotic confidence regions based on a broader family of test statistics known as power-divergence test statistics. In this study, we extend their work and propose penalized power-divergence test statistics-based confidence regions. We only consider small sample sizes where asymptotic properties fail and alternative methods are needed. Both bootstrap and asymptotic confidence regions are constructed. We consider the percentile and the bias corrected and accelerated bootstrap confidence regions. The latter confidence region has not been studied previously for the power-divergence statistics much less for the penalized ones. Designed simulation studies are carried out to calculate average coverage probabilities. Mean absolute deviation between actual and nominal coverage probabilities is used to compare the proposed confidence regions.     

Conditional estimation using prior information in 2‐stage group sequential designs assuming asymptotic normality when the trial terminated early

Two‐stage designs are widely used to determine whether a clinical trial should be terminated early. In such trials, a maximum likelihood estimate is often adopted to describe the difference in efficacy between the experimental and reference treatments; however, this method is known to display conditional bias. To reduce such bias, a conditional mean‐adjusted estimator (CMAE) has been proposed, although the remaining bias may be nonnegligible when a trial is stopped for efficacy at the interim analysis. We propose a new estimator for adjusting the conditional bias of the treatment effect by extending the idea of the CMAE. This estimator is calculated by weighting the maximum likelihood estimate obtained at the interim analysis and the effect size prespecified when calculating the sample size. We evaluate the performance of the proposed estimator through analytical and simulation studies in various settings in which a trial is stopped for efficacy or futility at the interim analysis. We find that the conditional bias of the proposed estimator is smaller than that of the CMAE when the information time at the interim analysis is small. In addition, the mean‐squared error of the proposed estimator is also smaller than that of the CMAE. In conclusion, we recommend the use of the proposed estimator for trials that are terminated early for efficacy or futility.     

Simultaneous confidence interval for assessing non‐inferiority with assay sensitivity in a three‐arm trial with binary endpoints

A three‐arm trial including an experimental treatment, an active reference treatment and a placebo is often used to assess the non‐inferiority (NI) with assay sensitivity of an experimental treatment. Various hypothesis‐test‐based approaches via a fraction or pre‐specified margin have been proposed to assess the NI with assay sensitivity in a three‐arm trial. There is little work done on confidence interval in a three‐arm trial. This paper develops a hybrid approach to construct simultaneous confidence interval for assessing NI and assay sensitivity in a three‐arm trial. For comparison, we present normal‐approximation‐based and bootstrap‐resampling‐based simultaneous confidence intervals. Simulation studies evidence that the hybrid approach with the Wilson score statistic performs better than other approaches in terms of empirical coverage probability and mesial‐non‐coverage probability. An example is used to illustrate the proposed approaches.     

Adaptive truncated weighting for improving marginal structural model estimation of treatment effects informally censored by subsequent therapy

Randomized clinical trials are designed to estimate the direct effect of a treatment by randomly assigning patients to receive either treatment or control. However, in some trials, patients who discontinued their initial randomized treatment are allowed to switch to another treatment. Therefore, the direct treatment effect of interest may be confounded by subsequent treatment. Moreover, the decision on whether to initiate a second‐line treatment is typically made based on time‐dependent factors that may be affected by prior treatment history. Due to these time‐dependent confounders, traditional time‐dependent Cox models may produce biased estimators of the direct treatment effect. Marginal structural models (MSMs) have been applied to estimate causal treatment effects even in the presence of time‐dependent confounders. However, the occurrence of extremely large weights can inflate the variance of the MSM estimators. In this article, we proposed a new method for estimating weights in MSMs by adaptively truncating the longitudinal inverse probabilities. This method provides balance in the bias variance trade‐off when large weights are inevitable, without the ad hoc removal of selected observations. We conducted simulation studies to explore the performance of different methods by comparing bias, standard deviation, confidence interval coverage rates, and mean square error under various scenarios. We also applied these methods to a randomized, open‐label, phase III study of patients with nonsquamous non‐small cell lung cancer. Copyright © 2015 John Wiley & Sons, Ltd.     

Corrected profile likelihood confidence interval for binomial paired incomplete data

Clinical trials often use paired binomial data as their clinical endpoint. The confidence interval is frequently used to estimate the treatment performance. Tang et al. (2009) have proposed exact and approximate unconditional methods for constructing a confidence interval in the presence of incomplete paired binary data. The approach proposed by Tang et al. can be overly conservative with large expected confidence interval width (ECIW) in some situations. We propose a profile likelihood‐based method with a Jeffreys' prior correction to construct the confidence interval. This approach generates confidence interval with a much better coverage probability and shorter ECIWs. The performances of the method along with the corrections are demonstrated through extensive simulation. Finally, three real world data sets are analyzed by all the methods. Statistical Analysis System (SAS) codes to execute the profile likelihood‐based methods are also presented. Copyright © 2013 John Wiley & Sons, Ltd.     

Selection bias for treatments with positive Phase 2 results

In drug development, treatments are most often selected at Phase 2 for further development when an initial trial of a new treatment produces a result that is considered positive. This selection due to a positive result means, however, that an estimator of the treatment effect, which does not take account of the selection is likely to over‐estimate the true treatment effect (ie, will be biased). This bias can be large and researchers may face a disappointingly lower estimated treatment effect in further trials. In this paper, we review a number of methods that have been proposed to correct for this bias and introduce three new methods. We present results from applying the various methods to two examples and consider extensions of the examples. We assess the methods with respect to bias of estimation of the treatment effect and compare the probabilities that a bias‐corrected treatment effect estimate will exceed a decision threshold. Following previous work, we also compare average power for the situation where a Phase 3 trial is launched given that the bias‐corrected observed Phase 2 treatment effect exceeds a launch threshold. Finally, we discuss our findings and potential application of the bias correction methods.     

Maximum-likelihood estimation for the removal method

We prove that the profile log-likelihood function for the removal method of estimating population size is unimodal. The result is obtained by a variation-diminishing property of the Laplace transform. An implication of this result is that the likelihood-ratio confidence region for the population size is always an interval. Necessary and sufficient conditions for the existence of a finite maximum-likelihood estimator are presented. We also present evidence that the likelihood-ratio confidence interval for the population size has acceptable small-sample coverage properties.     

Calibration Intervals in Linear Regression Models

Many of the existing methods of finding calibration intervals in simple linear regression rely on the inversion of prediction limits. In this article, we propose an alternative procedure which involves two stages. In the first stage, we find a confidence interval for the value of the explanatory variable which corresponds to the given future value of the response. In the second stage, we enlarge the confidence interval found in the first stage to form a confidence interval called, calibration interval, for the value of the explanatory variable which corresponds to the theoretical mean value of the future observation. In finding the confidence interval in the first stage, we have used the method based on hypothesis testing and percentile bootstrap. When the errors are normally distributed, the coverage probability of resulting calibration interval based on hypothesis testing is comparable to that of the classical calibration interval. In the case of non normal errors, the coverage probability of the calibration interval based on hypothesis testing is much closer to the target value than that of the calibration interval based on percentile bootstrap.     

Estimating confidence regions of common measures of the baseline and treatment effect on dichotomous outcome of a population

In this article, we estimate confidence regions of the common measures of (baseline, treatment effect) in observational studies, where the measure of a baseline is baseline risk or baseline odds, while the measure of a treatment effect is odds ratio, risk difference, risk ratio or attributable fraction, and where confounding is controlled in estimation of both the baseline and treatment effect. We use only one logistic model to generate approximate distributions of the maximum-likelihood estimates of these measures and thus obtain the maximum-likelihood-based confidence regions for these measures. The method is presented via a real medical example.     

Estimation Bias in the First-Order Autoregressive Model and Its Impact on Predictions and Prediction Intervals

The least squares estimate of the autoregressive coefficient in the AR(1) model is known to be biased towards zero, especially for parameters close to the stationarity boundary. Several methods for correcting the autoregressive parameter estimate for the bias have been suggested. Using simulations, we study the bias and the mean square error of the least squares estimate and the bias-corrections proposed by Kendall and Quenouille.

We also study the mean square forecast error and the coverage of the 95% prediction interval when using the biased least squares estimate or one of its bias-corrected versions. We find that the estimation bias matters little for point forecasts, but that it affects the coverage of the prediction intervals. Prediction intervals for forecasts more than one step ahead, when calculated with the biased least squares estimate, are too narrow.     

Asymptotic results in partially non-regular log-exponential distributions

We propose approximations to the moments, different possibilities for the limiting distributions and approximate confidence intervals for the maximum-likelihood estimator of a given parametric function when sampling from partially non-regular log-exponential models. Our results are applicable to the two-parameter exponential, power-function and Pareto distribution. Asymptotic confidence intervals for quartiles in several Pareto models have been simulated. These are compared to asymptotic intervals based on sample quartiles. Our intervals are superior since we get shorter intervals with similar coverage probability. This superiority is even assessed probabilistically. Applications to real data are included.