Optimum Allocation in Multivariate Stratified Random Sampling: A Modified Prékopa’s Approach

Considering the possible correlation between the characteristics (variables) in multivariate stratified random sampling, a modified Prékopa’s approach is suggested for the problem of optimum allocation in multivariate stratified random sampling. An example is solved by applying the proposed methodology.     

An Optimum Multivariate Stratified Sampling Design with Nonresponse: A Lexicographic Goal Programming Approach

In a multivariate stratified sample survey with L strata and p > 1 characteristics, defined on each unit of the population, let the estimation of all the p-population means be of interest. As discussed by Cochran (1977), since the optimum allocation for one characteristic will not in general be optimum for other characteristics some compromise must be reached in a multiple characteristics stratified surveys. Various authors worked out allocations that are based on a compromise criterion. The resulting allocations are optimal for all characteristics in some sense, for example an allocation that minimizes the trace of the variance-covariance matrix of the estimators of the population means or an allocation that minimizes the weighted average of the variances or an allocation that maximizes the total relative efficiency of the estimators as compared to the corresponding individual optimum allocations. In the present paper the problem of optimum allocation in multivariate stratified random sampling in the presence of nonresponse has been formulated as a multiobjective integer nonlinear programming problem and a solution procedure is developed using goal programming technique. Three numerical examples are worked out to illustrate the computational details. A comparison of the proposed method with some well known methods is also carried out to show the practical utility of the proposed method.     

An Optimal Chance Constraint Multivariate Stratified Sampling Design Using Auxiliary Information

When we are dealing with multivariate problem then we need an allocation which is optimal for all the characteristics in some sense because the individual optimum allocations usually differ widely unless the characteristics are highly correlated. So an allocation called “Compromise allocation” is to be worked out suggested by Cochran. When auxiliary information is also available, it is customary to use it to increase the precision of the estimates. Moreover, for practical implementation of an allocation, we need integer values of the sample sizes. In the present paper the problem is to determine the integer optimum compromise allocation when the population means of various characteristics are of interest and auxiliary information is available for the separate and combined ratio and regression estimates. This paper considers the optimum compromise allocation in multivariate stratified sampling with non-linear objective function and probabilistic non-linear cost constraint. The probabilistic non-linear cost constraint is converted into equivalent deterministic one by using Chance Constrained programming. The formulated multi-objective nonlinear programming problem is solved by Fuzzy Goal programming approach and Chebyshev approximation. Numerical illustration is also given to show the practical utility of the approaches.     

An optimum multivariate stratified double sampling design in presence of non-response

In stratified sampling when strata weights are unknown double sampling technique may be used to estimate them. At first a large simple random sample from the population without considering the stratification is drawn and sampled units belonging to each stratum are recorded to estimate the unknown strata weights. A stratified random sample is then obtained comprising of simple random subsamples out of the previously selected units of the strata. If the problem of non-response is there, then these subsamples may be divided into classes of respondents and non-respondents. A second subsample is then drawn out of non-respondents and an attempt is made to obtain the information. This procedure is called Double Sampling for Stratification (DSS). Okafor (Aligarh J Statist 14:13–23, 1994) derived DSS estimators based on the subsampling of non-respondents. Najmussehar and Bari (Aligarh J Statist 22:27–41, 2002) discussed an optimum double sampling design by formulating the problem as a mathematical programming problem and used the dynamic programming technique to solve it. In the present paper a multivariate stratified population is considered with unknown strata weights and an optimum sampling design is proposed in the presence of non-response to estimate the unknown population means using DSS strategy. The problem turns out to be a multiobjective integer nonlinear programming problem. A solution procedure is developed using Goal Programming technique. A numerical example is presented to illustrate the computational details.     

Minimum Cost Compromise Mixed Allocation

When more than one (say p) characteristics in multivariate stratified population are defined on each unit of the population, the individual optimum allocations may differ widely and can not be used practically. Moreover, there may be a situation such that no standard allocation is advisable to all the strata, for one reason or another. In such a situation, Clark and Steel (J R Stat Soc, Ser D Stat 49(2):197–207, 2000) suggested that different allocations may be used for different groups of strata having some common characteristics for double sampling in stratification. Later on, Ahsan et al. (Aligarh J Stat 25:87–97, 2005) used the same concept in univariate stratified sampling. They minimized the variance of the stratified sample mean for a fixed cost to obtain an allocation and called this allocation “mixed allocation”. In the present paper, a “compromise mixed allocation” is worked out for the fixed precisions of the estimates of the p-population means of a multivariate stratified population. A numerical example is also presented.     

Estimation of more than one parameters in stratified sampling with fixed budget

In a multivariate stratified sampling more than one characteristic are defined on every unit of the population. An optimum allocation which is optimum for one characteristic will generally be far from optimum for others. A compromise criterion is needed to work out a usable allocation which is optimum, in some sense, for all the characteristics. When auxiliary information is also available the precision of the estimates of the parameters can be increased by using it. Furthermore, if the travel cost within the strata to approach the units selected in the sample is significant the cost function remains no more linear. In this paper an attempt has been made to obtain a compromise allocation based on minimization of individual coefficients of variation of the estimates of various characteristics, using auxiliary information and a nonlinear cost function with fixed budget. A new compromise criterion is suggested. The problem is formulated as a multiobjective all integer nonlinear programming problem. A solution procedure is also developed using goal programming technique.     

Compromise allocation in multivariate stratified sample surveys under two stage randomized response model

A general model for the randomized response (RR) method was introduced by Warner (J. Am. Stat. Assoc. 60:63–69, 1965) when a single-sensitive question is under study. However, since social surveys are often based on questionnaires containing more than one sensitive question, the analysis of multiple RR data is of considerable interest. In multivariate stratified surveys with multiple RR data the choice of optimum sample sizes from various strata may be viewed as a multiobjective nonlinear programming problem. The allocation thus obtained may be called a “compromise allocation” in sampling literature. This paper deals with the two-stage stratified Warner’s RR model applied to multiple sensitive questions. The problems of obtaining compromise allocations are formulated as multi-objective integer non linear programming problems with linear and quadratic cost functions as two separate problems. The solution to the formulated problems are achieved through goal programming technique. Numerical examples are presented to illustrate the computational details.     

A biobjective method for sample allocation in stratified sampling

The two main and contradicting criteria guiding sampling design are accuracy of estimators and sampling costs. In stratified random sampling, the sample size must be allocated to strata in order to optimize both objectives.     

Adaptive Optimal Allocation in Stratified Sampling Methods

In this paper, we propose a stratified sampling algorithm in which the random drawings made in the strata to compute the expectation of interest are also used to adaptively modify the proportion of further drawings in each stratum. These proportions converge to the optimal allocation in terms of variance reduction and our stratified estimator is asymptotically normal with asymptotic variance equal to the minimal one. Numerical experiments confirm the efficiency of our algorithm. For the pricing of arithmetic average Asian options in the Black and Scholes model, the variance is divided by a factor going from 1.1 to 50.4 (depending on the option type and the moneyness) in comparison with the standard allocation procedure, while the increase in computation time does not overcome 1%.     

Estimation of some functional of the population distribution based on a stratified random sample

Summary In the preceding papers ([7], [8] and [9]), one of the authors discussed about the estimation of variances, covariances and correlation coefficients of the population based on a stratified random sample. In this paper we consider more general problem; estimating some functional θ(F) of the population distributionF based on a stratified random sample, which include our previous papers as special cases. We propose an unbiased estimator of θ(F) based on a stratified random sample and give an asymptotic expression of the gain in precision due to stratification in the case of proportional allocation. Furthermore, we present the general form of the optimum stratification in the proportional allocation for the estimation of θ(F).     

On stability in multiobjective programming — A stochastic approach

In this paper we assume that a deterministic multiobjective programming problem is approximated by surrogate problems based on estimations for the objective functions and the constraints. Making use of a large deviations approach, we investigate the behaviour of the constraint sets, the sets of efficient points and the solution sets if the size of the underlying sample tends to infinity. The results are illustrated by applying them to stochastic programming with chance constraints, where (i) the distribution function of the random variable is estimated by the empirical distribution function, (ii) certain parameters have to be estimated.     

Comparison of independent,stratified and random covering sample schemes in optimization problems

We consider two schemes of global optimization algorithms based on the use of grids. Our main goal is to compare the so-called independent sampling (IS), stratified sampling (SS) and random covering (RC) grids implemented to the estimation problem of the global maximum of a function. The results give an insight on how a decrease of randomness in selection rules for the trial points improves efficiency of global random search algorithms.     

Nonlinear integer programming for optimal allocation in stratified sampling

A stratified random sampling plan is one in which the elements of the population are first divided into nonoverlapping groups, and then a simple random sample is selected from each group. In this paper, we focus on determining the optimal sample size of each group. We show that various versions of this problem can be transformed into a particular nonlinear program with a convex objective function, a single linear constraint, and bounded variables. Two branch and bound algorithms are presented for solving the problem. The first algorithm solves the transformed subproblems in the branch and bound tree using a variable pegging procedure. The second algorithm solves the subproblems by performing a search to identify the optimal Lagrange multiplier of the single constraint. We also present linearization and dynamic programming methods that can be used for solving the stratified sampling problem. Computational testing indicates that the pegging branch and bound algorithm is fastest for some classes of problems, and the linearization method is fastest for other classes of problems.     

A single-resource allocation problem with Poisson resource requirements

We consider a stochastic resource allocation problem that generalizes the knapsack problem to account for random item weights that follow a Poisson distribution. When the sum of realized weights exceeds capacity, a penalty cost is incurred. We wish to select the items that maximize expected profit. We provide an effective solution method and illustrate the advantages of this approach via computational experiments.     

Computing the Pareto frontier of a bi-objective bi-level linear problem using a multiobjective mixed-integer programming algorithm

In this article, we study the bi-level linear programming problem with multiple objective functions on the upper level (with particular focus on the bi-objective case) and a single objective function on the lower level. We have restricted our attention to this type of problem because the consideration of several objectives at the lower level raises additional issues for the bi-level decision process resulting from the difficulty of anticipating a decision from the lower level decision maker. We examine some properties of the problem and propose a methodological approach based on the reformulation of the problem as a multiobjective mixed 0–1 linear programming problem. The basic idea consists in applying a reference point algorithm that has been originally developed as an interactive procedure for multiobjective mixed-integer programming. This approach further enables characterization of the whole Pareto frontier in the bi-objective case. Two illustrative numerical examples are included to show the viability of the proposed methodology.     

Allocation of dependent divisional resources

A resource allocation problem is considered with resources that are dependent in the sense that an allocation to an activity requires the application of several resources, except for certain activities which are divisional in the sense that an allocation to such an activity requires the use of only a single resource. Return and cost functions are assumed to be continuous and increasing, and the allocation variables are continuous. Conditions are given for the replacement of the continuous problem by an associated problem with discrete variables and a single constraint, and to a given degree of accuracy. The associated problem can be efficiently solved by dynamic programming. Certain divisional resource allocation problems with discrete variables and several linear constraints are shown to be equivalent to a discrete problem with a single constraint. A numerical example is given.     

An arc-exchange decomposition method for multistage dynamic networks with random arc capacities

Multistage dynamic networks with random arc capacities (MDNRAC) have been successfully used for modeling various resource allocation problems in the transportation area. However, solving these problems is generally computationally intensive, and there is still a need to develop more efficient solution approaches. In this paper, we propose a new heuristic approach that solves the MDNRAC problem by decomposing the network at each stage into a series of subproblems with tree structures. Each subproblem can be solved efficiently. The main advantage is that this approach provides an efficient computational device to handle the large-scale problem instances with fairly good solution quality. We show that the objective value obtained from this decomposition approach is an upper bound for that of the MDNRAC problem. Numerical results demonstrate that our proposed approach works very well.     

A Heuristic for Moment-Matching Scenario Generation

In stochastic programming models we always face the problem of how to represent the random variables. This is particularly difficult with multidimensional distributions. We present an algorithm that produces a discrete joint distribution consistent with specified values of the first four marginal moments and correlations. The joint distribution is constructed by decomposing the multivariate problem into univariate ones, and using an iterative procedure that combines simulation, Cholesky decomposition and various transformations to achieve the correct correlations without changing the marginal moments.With the algorithm, we can generate 1000 one-period scenarios for 12 random variables in 16 seconds, and for 20 random variables in 48 seconds, on a Pentium III machine.     

Stratified two-stage sampling in domains: Sample allocation between domains, strata, and sampling stages

In the paper, formulae for optimum sample allocation between domains, strata in the domains, and sampling stages are presented for stratified two-stage sampling in domains under fixed sample size of SSUs from PSUs.     

Partitioned Random Search for Global Optimization with Sampling Cost and Discounting Factor

The method of partitioned random search has been proposed in recent years to obtain an as good as possible solution for the global optimization problem (1). A practical algorithm has been developed and applied to real-life problems. However, the design of this algorithm was based mainly on intuition. The theoretical foundation of the method is an important issue in the development of efficient algorithms for such problems. In this paper, we generalize previous theoretical results and propose a sequential sampling policy for the partitioned random search for global optimization with sampling cost and discounting factor. A proof of the optimality of the proposed sequential sampling policy is given by using the theory of optimal stopping.