Abstract
We construct an optimal design to simultaneously estimate three common interesting features in a dose-finding trial with possibly different emphasis on each feature. These features are (1) the shape of the dose-response curve, (2) the median effective dose and (3) the minimum effective dose level. A main difficulty of this task is that an optimal design for a single objective may not perform well for other objectives. There are optimal designs for dual objectives in the literature but we were unable to find optimal designs for 3 or more objectives to date with a concrete application. A reason for this is that the approach for finding a dual-objective optimal design does not work well for a 3 or more multiple-objective design problem.
We propose a method for finding multiple-objective optimal designs that estimate the three features with user-specified higher efficiencies for the more important objectives. We use the flexible 4-parameter logistic model to illustrate the methodology but our approach is applicable to find multiple-objective optimal designs for other types of objectives and models. We also investigate robustness properties of multiple-objective optimal designs to mis-specification in the nominal parameter values and to a variation in the optimality criterion. We also provide computer code for generating tailor made multiple-objective optimal designs.
1 Background
Experiments are increasingly expensive to conduct and it is desirable to obtain maximal information at minimal cost. Researchers now want to have several research questions answered from a single study to save cost. Designing such multi-objective experiments can be challenging because an optimal design for one objective can perform poorly under another and further, not all objectives may be equally important. Contrary to a single-objective optimal design, the sought design has to incorporate the multiple objectives at the onset and provide user-specified efficiencies for making inferences for the more important objectives.
There is some work on constructing dual-objective optimal designs but there is little in the literature on construction and properties of optimal designs for 3 or more objectives with a concrete application. Our aims in this paper are to construct 3-objective optimal designs for a pharmaceutical application, investigate effectiveness of such designs over single-objective optimal designs, and study their robustness properties under mis-specification in the nominal values of the model parameters and a change in the optimality criterion. Our focus is to estimate interesting characteristics of an agent in a dose response study where some drug characteristics may be more meaningful or important than others. Our dose response model is the flexible 4-parameter logistic model widely used in many disciplines, such as in educational research, biological sciences, pharmaceutical sciences and agronomy, to name a few. Both our application and the model are illustrative in the sense that the methodology described here also applies to other models and criteria.
Our setup assumes that we have a nonlinear regression model defined on a given compact dose interval X. The model has a known mean structure with unknown parameters and errors are assumed to be normal, independent and identically distributed, each with mean 0 and constant variance
We follow convention and measure the worth of a design by its information matrix defined as the negative of the expectation of the second derivatives of the log likelihood function with respect to the model parameters. Given an approximate design
Section 2 describes the methodology for searching optimal designs when there are 3 or more objectives of varying interests. In Section 3, we discuss optimality criteria for dose response studies in the context of the popular Hill model and its relationship with the 4-parameter logistic model. In Section 4, we revisit a dose response study with 7 anti-cancer drugs from Khinkis et al. [4], report the single-objective optimal designs for estimating all parameters in the mean function, the
2 Multiple-objective optimal designs
Multiple-objective optimal designs are appealing because many scientific studies have several objectives and these objectives may vary in importance. A properly constructed design allocates resources that ensures the more important objectives are attained with user-specified efficiencies at minimum cost. Such situations arise frequently in real studies. We give three examples. In a dose response study, there is interest to infer the mean response at a specific dose and to estimate the shape of the overall dose curve accurately, and one of these objectives may be more important than the other. Another example is in estimation problems. It is often the case that some parameters are more meaningful than others and so there is greater interest in estimating selected parameters more accurately. For instance, in the 2-parameter Michaelis-Menten model, the Michaelis-Menten constant is clearly more interesting to estimate than the other parameter because it controls the rate of an enzyme-kinetic reaction. The third example concerns model inadequacy and inference. A properly targeted multiple-objective optimal design can detect model inadequacies and provide accurate inference at the same time. For example, consider the Emax model, which is the same as the Michaelis-Menten model except that the substrate concentration variable in the model is raised to some power. With the power as the third parameter, the Emax model is more flexible and can better capture asymmetry in the mean dose response curve than the Michaelis-Menten model. Model inadequacy concern and accurate inference on the parameters for the Emax model can then be simultaneously incorporated at the design stage using a multiple-objective optimal design that requires the power parameter be estimated with a user-specified efficiency, say
Multiple-objective optimal designs date back to the early seventies, where the proposed methods for constructing such designs were largely either based on ad-hoc procedures or simply based on the hope that the design constructed for the most important objective will be adequate for the other objectives. Stigler [5], Lauter [6, 7] and Lee [8, 9] were early attempts to formalize the procedure. Most were concerned with polynomial regression problems. For instance, Studden [10] was concerned about model inadequacy and wanted to find a design that was robust to the degree of the assumed polynomial model. His method ensured that coefficients in the assumed model were estimated as accurately as possible, and at the same time, the design could also provide user-specified efficiencies for estimating coefficients that might be needed for a higher degree polynomial model. Subsequent work on finding dual-objectives optimal designs includes Dette [11, 12], Zhu et al. [13], Wong [14], Song and Wong [15], Tsai and Zen [16], Atkinson [17], McGree et al. [18], Tommasi [19] and, Padmanabhan and Dragalin [20]. A recent application is Zhang et al. [21] where they constructed dual-objective optimal designs for a mixture experiment.
The formulation of the multiple-objective optimal design problem invariably involves a constrained optimization problem where the goal is to find a design that simultaneously meets user-specified minimal efficiencies for the various criteria, with higher efficiencies sought for the more important criteria, and subject to these requirements, does as well as possible for the least important criterion. Of course the sought multiple-objective optimal design may not exist if the requirements are too stringent and the objectives are competitive, meaning that much efficiency of one type has to be given up for a small gain in another criterion. Ad-hoc methods generally seek to combine the multiple design criteria into a single criterion with the expectation that the resulting optimal design for the single combined criterion may be efficient for all the criteria. The common problems with such an approach include how to combine the criteria in a meaningful way and the unclear interpretation of the combined criterion.
A more formal method to search for multiple-objective optimal approximate designs when there are two objectives
Clyde and Chaloner [24] extended the methodology to find Bayesian multiple-objective optimal designs for nonlinear models. However, there is no work to date that focuses on constructing an optimal design for 3 or more different objectives with a concrete application. We were also unable to find work that studies robustness properties of multiple-objective optimal designs to model mis-specifications or under a change in the optimality criterion. There are only a handful of related papers that either briefly considered finding an optimal design for a problem with 3 or more objectives or they addressed a different class of problems. For example, El-Monsef and Seyam [25] proposed optimal designs for model discrimination, parameter estimation, and estimation of a function of parameters. The paper combined the optimality criteria for the three objectives using a weighted compound criterion. However, despite the title of their paper, only the last half page outlined how one may construct a specific 3-objective optimal design. There were no practical details, examples and explanation on how to meaningfully select the weights in the combined design criterion for maximizing each efficiency. Another related work is Antognini and Zagoraiou [26], who considered a non dose-response setup and their goal was to determine an adaptive optimal allocation scheme for subjects to treatment groups under 3 criteria in a clinical trial. They wanted to balance the competing needs of ethics requirements, proper randomization and precise treatment efficacy estimation. In such problems, the number of treatment groups is fixed (i.e. the design space consists only of a few points) and so only the optimal proportion or optimal number of subjects to assign to each treatment has to be determined. Our optimization problems have a continuous dose interval and we need to determine the number of optimal dose levels, the optimal dose levels, and the optimal proportions or numbers of subjects to assign at the dose levels. Our constrained optimization problems are thus more difficult because they have more variables to optimize over a continuous multi-dimensional space. Earlier, Zhu and Wong [27] also found multiple-objective optimal designs but their setup was limiting; their interest was confined only to estimating percentiles in a two parameter logistic model and did not discuss different types of criteria, which we have here.
The dearth of work for finding an optimal design for 3 or more objectives in a dose response study with a concrete application can be partially explained by the complexity of the efficiency plot, which increases in dimension as the number of objectives increases. In a high-dimensional efficiency plot, the visual appreciation of the shapes of various efficiency plots becomes compromised and it becomes difficult to identify the correct vector
This paper presents a systematic approach to construct a multiple-objective optimal design for the 3 common objectives in a dose response study and the methodology can be directly applied to find other types of multiple-objective optimal designs in other problems. We provide an efficient algorithm for searching the multiple-objective optimal design that meets different user-specified efficiencies for the objectives and for evaluating efficiency of the generated design under various criteria. We also provide a concrete application to a dose response study and study robustness properties of the multiple-objective optimal design to mis-specification in nominal values for the model parameters and under a change of criterion. Bayesian multiple-objective optimal designs can also be found using our approach and we provide an example of such an optimal design with 10 dose levels found from our algorithm.
3 Objectives, models and algorithms for finding optimal designs in dose response studies
We assume that the continuous response variable from the
where
3.1 Common objectives in dose response models
Some interesting characteristics of a drug are the shape of the dose-response, the median effective dose (
A D-optimal design minimizes the volume of the ellipsoidal confidence region of the model parameters and so estimates from the D-optimal design are the most precise. When the goal is to estimate a function of the model parameters, such as the
3.2 A common dose response model
We now consider a versatile and popular model commonly used in dose response studies and several other disciplines. The 4-parameter Hill model has a continuous outcome and its mean response is given by
Here
Khinkis et al. [4] conducted a cell growth inhibition study in a laboratory to investigate the effectiveness of 7 anticancer drugs to shrink the tumor using the 4-parameter Hill model. Nominal parameter values for these drugs used in the study are displayed in Table 1. Figure 1 shows the mean response shapes of (2) for the 7 different sets of values of the parameters in Khinkis et al. [4], suggesting that the characteristics of the drugs are quite different.
Drug | Drug | ||
TMTX | AG2009 | ||
MTX | AG337 | ||
AG2034 | ZD1694 | ||
AG2032 |
Let
which is sometimes referred to as the 4-parameter logistic model. This form was used in Li and Majumdar [31] and is equivalent to the above form with
The 4-parameter logistic model has an advantage over the Hill model in that it tends to provide more stable parameter estimates [34]. In what is to follow, we use the 4-parameter logistic model (2) and redesign the study in Khinkis et al. [4] using optimal design theory and compare benefits of our locally multiple-objective optimal designs with their locally D-optimal designs.
If the approximate design
and
When estimates
4 Dose response optimal designs
We now present single-objective optimal designs, multiple-objective optimal designs and investigate robustness properties of the latter designs to mis-specification in nominal parameter values and changes in the objectives. The nominal values in Table 1 were used to construct optimal designs on the dose interval
All optimal designs in this paper were found based on the Yang-Biedermann-Tang (YBT) algorithm that has been shown to converge to an optimal design for a large class of design problems [36]. The authors also used several examples and showed that their algorithm performed faster than current algorithms for finding single-objective optimal designs, including a traditional class of algorithms such as the V-algorithm. The YBT algorithm requires that the dose range to be discretized. If there are s parameters in the mean function, the algorithm uses a starting design with
We discovered that there was a problem in the YBT algorithm when we applied it to search for multiple-objective optimal designs. If the randomly selected
4.1 Single-objective optimal designs
We recall that a D-optimal design
where
Similar to estimating the
where
In the rest of the paper, we assume
Table 2 displays the single-objective optimal designs found from our modified algorithm. All optimal designs found by the algorithm are verified by an Equivalence Theorem. This is an important tool in optimal design theory that enables one to confirm if a design is optimal among all designs on the given dose interval. It is derived from the Frechet derivative of the convex(or concave) optimality criterion, and while all the Equivalence Theorems have similar forms, each criterion has its own Equivalence Theorem. For example, for D-optimality, the Equivalence Theorem states that if we have a homoscedastic model and the mean response function has s parameters, the design
for all dose levels x in the dose interval X, with equality when x is a dose level of the design
Equivalence Theorems for other types of optimal designs are available in design monographs, see for example, Pukelsheim [39] and Atkinson et al. [40]. For example, in order to verify if a design
or
In the literature, the function on the left hand side of the inequality is sometimes refereed to as the sensitivity function. Figure 2 shows the plot of the sensitivity function for each of the single-objective optimal designs when
Table 2 shows the optimal designs for different nominal sets for
4.2 Multiple-objective optimal designs
We now apply our modified algorithm to search for multiple-objective optimal designs for estimating
The efficiencies of using
and
The above measures are all between 0 and 1 and have the following interpretation: if
Here each
Each summand in (6) has been properly scaled, and it is easy to see that if all but one of the weights is nonzero, (6) reduces to the Equivalence Theorem for the single-objective optimal design. The Equivalence Theorem states that for a given vector
with equality when x is a dose level of the design
All multiple-objective optimal designs were obtained using a modified version of the YBT algorithm and confirmed using the above Equivalence Theorem in (6). As noted earlier, without the modification, we were unable to find the multiple-objective optimal designs for some cases. One reason appears to be that the YBT algorithm sometimes begins its search using poor initial dose levels that are far from the optimal dose levels. The modification we made was to first run the V-algorithm r times using the sensitivity function (6) and then select the last
Example 1: We used the set of nominal values
Example 2: Miller et al. [29] assumed
Table 3 shows the multiple-objective optimal designs from the YBT algorithm and our modified algorithm for the two examples. Figures 3 and 4 confirm the optimality of the two multiple-objective optimal designs found from our modified algorithm but not the designs found from the YBT algorithm.
Example 1 | |
YBT algorithm | |
Modified algorithm | |
Example 2 | |
YBT algorithm | |
Modified algorithm |
Table 4 shows the multiple-objective optimal designs for the different
Design | Design | ||||||||
1 | 0.599 | 0.511 | 1 | 0.583 | 0.505 | ||||
1 | 0.006 | 1 | 0 | ||||||
0 | 1 | 0.070 | 1 | ||||||
0.866 | 0.815 | 0.746 | 0.866 | 0.786 | 0.696 | ||||
1 | 0.602 | 0.480 | 1 | 0.596 | 0.474 | ||||
1 | 0.001 | 1 | 0.002 | ||||||
0 | 1 | 0 | 1 | ||||||
0.894 | 0.762 | 0.645 | 0.895 | 0.761 | 0.642 | ||||
1 | 0.595 | 0.493 | 1 | 0.595 | 0.470 | ||||
1 | 0 | 1 | 0.002 | ||||||
0.161 | 1 | 0.063 | 1 | ||||||
0.880 | 0.782 | 0.694 | 0.891 | 0.766 | 0.657 | ||||
1 | 0.682 | 0.478 | |||||||
1 | 0 | ||||||||
0.069 | 1 | ||||||||
0.895 | 0.761 | 0.650 |
How does a multiple-objective optimal design perform under a single-objective criterion? Table 5 displays these efficiencies and also shows how well single-objective optimal designs perform under a variation of criterion. As expected, single-objective optimal designs have efficiency 1 under their own criterion but do not perform well under different objectives. On the other hand, the multiple-objective optimal design
For multiple-objective optimal designs, the components in the weight vector
0.55 | 0.912 | 0.825 | |
0.90 | 0.995 | 0.653 | |
0.95 | 0.999 | 0.621 | |
1.00 | 1.000 | 0.596 |
For 3 or more-objectives, it is harder to visualize the changes in the efficiencies of the generated design as the weights vary in the compound criterion from the high-dimensional efficiency plot. To this end, we generated multiple-objective optimal designs for each possible pair of values for
In practice, practitioners first prioritize the importance of each objective and set efficiency requirements for each objective with higher efficiencies for the more important ones. For example, suppose
4.3 Robustness properties of multiple-objective optimal designs
This section investigates impact of uncertainty of the nominal parameter values on the multiple-objective optimal designs. This can arise when there are different single best guesses for the set of nominal values for the model parameters or there are several competing sets for the nominal values. In the latter case, one could adopt a Bayesian approach that averages the criterion over the various choices and optimize the resulting criterion. If nominal parameter values come from equally good previous studies or from equally qualified experts, one may use a uniform prior to average out the uncertainty; otherwise weights assigned to different sets of nominal parameter values may be chosen to reflect their plausibility. We now report the results from our investigation on how mis-specification in nominal values of the model parameters affect the performance of the multiple-objective optimal designs.
We first investigate whether multiple-objective optimal designs generally provide higher efficiencies for all criteria than single-objective optimal designs when there is uncertainty in the nominal parameter values. Consider the seven different sets of nominal values for
Here
where
1 | 1 | 1 | 0 | 0 | 0 | ||||||||
0.562 | 0.168 | 0.659 | 0.001 | 0 | 0 | ||||||||
0.007 | 0 | 0 | 0.025 | 0.022 | 0 | ||||||||
0.079 | 0.003 | 0 | 0.002 | 0 | 0 | ||||||||
0 | 0 | 0 | 1 | 1 | 1 | ||||||||
0.005 | 0.009 | 0 | 0.019 | 0 | 0.056 | ||||||||
0.280 | 0.016 | 0 | 0.001 | 0 | 0 | ||||||||
0.743 | 0.390 | 0.328 | 0.407 | 0.227 | 0.434 | ||||||||
0.418 | 0.018 | 0 | 0.011 | 0 | 0.007 | ||||||||
1 | 1 | 1 | 0.034 | 0 | 0.025 | ||||||||
0.002 | 0 | 0 | 0.783 | 0.003 | 0.220 | ||||||||
0.140 | 0.006 | 0 | 0.185 | 0.003 | 0 | ||||||||
0 | 0 | 0 | 0.002 | 0.003 | 0 | ||||||||
0.002 | 0.008 | 0 | 1 | 1 | 1 | ||||||||
0.551 | 0.036 | 0.780 | 0.117 | 0.001 | 0.220 | ||||||||
0.658 | 0.462 | 0.329 | 0.667 | 0.450 | 0.494 | ||||||||
0.099 | 0 | 0.322 | 0.300 | 0.013 | 0 | ||||||||
0.174 | 0.002 | 0 | 0.693 | 0.240 | 0.880 | ||||||||
1 | 1 | 1 | 0.113 | 0.003 | 0 | ||||||||
0.365 | 0.017 | 0 | 0.679 | 0.321 | 0.005 | ||||||||
0.050 | 0 | 0 | 0 | 0 | 0 | ||||||||
0.882 | 0.541 | 0.030 | 0.090 | 0.003 | 0 | ||||||||
0.360 | 0.016 | 0 | 1 | 1 | 1 | ||||||||
0.766 | 0.537 | 0.455 | 0.890 | 0.491 | 0.398 | ||||||||
0.005 | 0.958 | 0 | |||||||||||
0.071 | 0 | 0.004 | |||||||||||
0.017 | 0 | 0 | |||||||||||
1 | 1 | 1 | |||||||||||
0 | 0 | 0 | |||||||||||
0.014 | 0 | 0 | |||||||||||
0.458 | 0.015 | 0.829 | |||||||||||
0.732 | 0.443 | 0.418 |
Figure 8 shows that the sensitivity function of the robust multiple-objective optimal design has a maximum value of 0 at the optimal dose levels over the dose interval and so confirms the optimality of the generated design. Table 7 displays the efficiencies of
5 Summary
We presented a systematic method for finding multiple-objective optimal designs for the 4-parameter logistic regression model. The objectives were to estimate (1) the overall dose-response curve; (2) the
The YBT algorithm was recently proposed as a state-of-the-art algorithm to find single-objective optimal designs, but we found that the algorithm can sometimes fail to find multiple-objective optimal designs. We overcame the problem by modifying the YBT algorithm by ensuring it selects better initial dose levels using the traditional V-algorithm. Our modified algorithm finds tailor-made multiple-objective optimal designs and can sometimes also outperform the YBT algorithm for searching single-objective optimal designs. For instance, in Examples 1 and 2, the YBT algorithm took 7.80 and 4.91 s respectively, to find the D-optimal designs and our modified algorithm using
Funding statement: Funding: The research of Wong reported in this paper was partially supported by the U.S. Department of Health and Human Services, National Institute Of General Medical Sciences of the National Institutes of Health under Award Number R01GM107639.
Acknowledgement
The contents in this paper are solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
References
1. Kiefer J. Jack Carl Kiefer collected papers III, design of experiments. New York: Springer-Verlag, 1985.10.1007/978-1-4615-6660-1Search in Google Scholar
2. Pazman A. Some features of the optimal design theory-a survey. Ser Stat 1980;11:415–446.10.1080/02331888008801550Search in Google Scholar
3. Chernoff H. Locally optimal designs for estimating parameters. Ann Math Stat 1953;24:586–602.10.1214/aoms/1177728915Search in Google Scholar
4. Khinkis LA, Levasseur L, Faessel H, Greco WR. Optimal design for estimating parameters of the 4-parameter hill model. Nonlinearity Biol Toxicol Med 2003;1:363–377.10.1080/15401420390249925Search in Google Scholar PubMed PubMed Central
5. Stigler SM. Optimal experimental design for polynomial regression. J Am Stat Assoc 1971;66:311–318.10.1080/01621459.1971.10482260Search in Google Scholar
6. Lauter E. Experimental design in a class of models. Math Operationsforsch Stat 1974;5:379–398.10.1080/02331887408801175Search in Google Scholar
7. Lauter E. Optimal multipurpose designs for regression models. Math Operationsforsch Stat 1974;7:51–68.10.1080/02331887608801276Search in Google Scholar
8. Lee CM-S. Constrained optimal designs for regression models. Commun Stat Theory Methods 1987;16:765–783.10.1080/03610928708829401Search in Google Scholar
9. Lee CM-S. Constrained optimal designs. J Stat Plann Inference 1988;18:377–389.10.31274/rtd-180813-13130Search in Google Scholar
10. Studden WJ. Some robust-type D-optimal designs in polynomial regression. J Am Stat Assoc 1982;77:916–921.10.2307/2287327Search in Google Scholar
11. Dette H. Optimal designs for a class of polynomials of odd or even degree. Ann Stat 1992;20:238–259.10.1214/aos/1176348520Search in Google Scholar
12. Dette H. On a mixture of the D- and D1${D_1}$-optimality criterion in polynomial regression. J Stat Plann Inference 1993;35:233–249.10.1016/0378-3758(93)90047-ASearch in Google Scholar
13. Zhu W, Ahn H, Wong WK. Multiple-objective optimal designs for the logit model. Commun Stat Theory Methods 1998;27:1581–1592.10.1080/03610929808832178Search in Google Scholar
14. Wong WK. Recent advances in constrained optimal design strategies. Stat Neerl 1999;53:257–276.10.1111/1467-9574.00111Search in Google Scholar
15. Song D, Wong WK. On the construction of Grm-optimal designs. Stat Sin 1999;9:263–272.Search in Google Scholar
16. Tsai M, Zen M. Criterion-robust optimal designs for model discrimination and parameter estimation: multivariate polynomial regression case. Stat Sin 2004;14:591–601.Search in Google Scholar
17. Atkinson AC. DT-optimum designs for model discrimination and parameter estimation. J Stat Plann Inference 2008;1:56–64.10.1016/j.jspi.2007.05.024Search in Google Scholar
18. McGree JM, Eccleston JA, Duffull SB. Compound optimal design criteria for nonlinear models. J Biopharm Stat 2008;18:646–661.10.1080/10543400802071352Search in Google Scholar PubMed
19. Tommasi C. Optimal designs for both model discrimination and parameter estimation. J Stat Plann Inference 2009;139:4123–4132.10.1016/j.jspi.2009.05.042Search in Google Scholar
20. Padmanabhan SK, Dragalin V. Adaptive Dc-optimal designs for dose finding based on a continuous efficacy endpoint. Biom J 2010;52:836–852.10.1002/bimj.200900214Search in Google Scholar PubMed
21. Zhang C, Wong WK, Peng H. Dual-objective optimal mixture designs. Aust N Z J Stat 2012;54:211–222.10.1111/j.1467-842X.2012.00670.xSearch in Google Scholar
22. Cook RD, Wong WK. On the equivalence of constrained and compound optimal designs. J Am Stat Ass 1994;89:687–692.10.1080/01621459.1994.10476794Search in Google Scholar
23. Wong WK. A graphical approach for the construction of constrained D- and L-optimal designs using efficiency plots. J Stat Comput Simul 1995;53:143–152.10.1080/00949659508811702Search in Google Scholar
24. Clyde M, Chaloner K. The equivalence of constrained and weighted designs in multiple objective design problems. J Am Stat Assoc 1996;91:1236–1244.10.1080/01621459.1996.10476993Search in Google Scholar
25. Abd El-Monsef MME, Seyam MM. CDT-optimum designs for model discrimination, parameter estimation and estimation of a parametric function. J Stat Plann Inference 2011;141:639–643.10.1016/j.jspi.2010.07.010Search in Google Scholar
26. Baldi Antognini A, Zagoraiou M. Multi-objective optimal designs in comparative clinical trials with covariates: the reinforced doubly adaptive biased coin design. Ann Stat 2012;40:1315–1345.10.1214/12-AOS1007Search in Google Scholar
27. Zhu W, Wong WK. Bayesian optimal designs for estimating a set of symmetric quantiles. Stat Med 2001;20:123–137.10.1002/1097-0258(20010115)20:1<123::AID-SIM643>3.0.CO;2-5Search in Google Scholar
28. Huang YC, Wong WK. Sequential construction of multiple-objective optimal designs. Biometrics 1998;54:1388–1396.10.2307/2533665Search in Google Scholar
29. Miller F, Guilbaud O, Dette H. Optimal designs for estimating the interesting part of a dose-effect curve. J Biopharm Stat 2007;17:1097–1115.10.1080/10543400701645140Search in Google Scholar
30. Sebaugh JL. Guidelines for accurate EC50/IC50 estimation. Pharm Stat 2011;10:128–134.10.1002/pst.426Search in Google Scholar
31. Li G, Majumdar D. D-optimal designs for logistic models with 3 and four parameters. J Stat Plann Inference 2008;138:1950–1959.10.1016/j.jspi.2007.07.010Search in Google Scholar
32. Bretz F, Dette H, Pinheiro J. Practical considerations for optimal designs in clinical dose finding studies. Stat Med 2010;29:731–742.10.1002/sim.3802Search in Google Scholar
33. Ritz C, Streibig JC. Bioassay analysis using R. J Stat Software 2005;12:1–22.10.18637/jss.v012.i05Search in Google Scholar
34. Reeve R, Turner JR. Pharmacodynamic models: parameterizing the hill equation, Michaelis-Menten, the logistic curve, and relationships among these models. J Biopharm Stat 2013;23:648–661.10.1080/10543406.2012.756496Search in Google Scholar PubMed
35. Yang M. On the de la Garza phenomenon. Ann Stat 2010;38:2499–2524.10.1214/09-AOS787Search in Google Scholar
36. Yang M, Biedermann S, Tang E. On optimal designs for nonlinear models: a general and efficient algorithm. J Am Stat Assoc 2013;108:1411–1420.10.1080/01621459.2013.806268Search in Google Scholar
37. Fedorov VV. Theory of optimal experiments (transl. and ed. by W. J. Studden and E. M. Klimko). New York: Academic, 1972.Search in Google Scholar
38. Hyun SW, Wong WK, Yang Y. “VNM: multiple objective optimal design for the 4 parameter logistic model.” R package version 2.0, http://CRAN.R-project.org/package=VNM, 2015.Search in Google Scholar
39. Pukelsheim F. Optimal design of experiment. Philadelphia, PA: Society for Industrial and Applied Mathematics (SIAM), 2006.10.1137/1.9780898719109Search in Google Scholar
40. Atkinson AC, Donev AN, Tobias RD. Optimum experimental designs with SAS. Oxford: Oxford University Press, 2007.Search in Google Scholar
41. Sverdlov O, Ryeznik Y, Wong WK. Efficient and ethical response-adaptive randomization designs for multi-arm clinical trials with Weibull time-to-event outcomes. J Biopharm Stat 2014;24:732–754.10.1080/10543406.2014.903261Search in Google Scholar PubMed
© 2015 Walter de Gruyter GmbH, Berlin/Boston