Multi-objective geometric programming approach in multivariate stratified sample survey with quadratic cost function

Ullah, Shafi; Agarwal, Ravi P.

doi:10.1080/00949655.2015.1086765

Cited by 2 publications

(1 citation statement)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Dual solution is obtained by using LINGO [87] Software for solving FSOCP. The optimum allocations of SS of R and NR for both the phases have been obtained by applying DS and primal-dual relationship theorem as described by [44], [52], [66], [69], [70] and [80]. The given technique is illustrated by projecting a numerical problem.…”

Section: Introductionmentioning

confidence: 99%

A New Fuzzy Multiobjective Geometric Programming in Double Sampling in Presence of Non-Response

Shafiullah¹,

Tariq

Bari

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Non-linear programming problems can be solved efficiently by geometric programming in which non-linear expressions can be expressed by an exponential or power functions. In this paper, Double Sampling Stratification (DSS) strategy is used in multivariate stratified population with unidentified strata weights, for optimum sampling design (OSD) in the presence of non-response to estimate the unidentified population means. The double sampling problem in the presence of non-response is formulated as Fuzzy Multiobjective Convex Programming Problem (FMOCPP). Then, we convert FMOCPP into Fuzzy Single Objective Convex Problem (FSOCP) by using membership function. Dual solution is obtained using LINGO Software by solving FSOCP. The optimum allocations of sample sizes of respondents and non-respondents for both the phases are obtained by applying dual solutions and primal-dual relationship theorem. The given technique is illustrated by projecting a numerical problem. INDEX TERMS Double sampling, dual solutions (DS), fuzzy programming (FP), geometric programming (GP), non-linear programming (NLP), non-response, non-respondents (NR), orthogonality condition (OC), respondents (R).

show abstract

Section: Introductionmentioning

confidence: 99%