In the stratified sampling literature the main problem is to determine the sample sizes that should be selected from each stratum under (i) Equal Allocation, (ii) Proportional Allocation and (iii) Optimum Allocation. The best method is optimum allocation but in real situation the implementation of optimum allocation is not possible. In this case it is of interest to find near optimal allocation or compromise allocation. In case of multivariate sampling problem (where p different characteristics are under study) the optimal allocation method does not give the optimal solution for each variable and then researcher have to adapt in solution up to some extent by which the solution gives the optimal allocation in some sense. The compromise allocation is advisable in this situation. The present paper discusses a real situation problem where the two stage sampling are under study for more than one characteristics with quadratic travel cost of survey, the problem can be formulated as Multivariate Non Linear Programming Problem (MNLPP). The MNLPP is then solved by Dynamic Programming Technique with a numerical example.

Citation details of the article

Journal: International Journal of Applied Mathematics
Journal ISSN (Print): ISSN 1311-1728
Journal ISSN (Electronic): ISSN 1314-8060
Volume: 35
Issue: 1
Year: 2022

DOI: 10.12732/ijam.v35i1.13

Download Section

Download the full text of article from here.

You will need Adobe Acrobat reader. For more information and free download of the reader, please follow this link.


  1. [1] A.H. Ansari, Compromise allocation for two stage cluster sampling using dynamic programming technique, International J. of Mathematical Archive, 6, No 4 (2015), 10–15.
  2. [2] B.F. Arnold, Procedures to determine optimum two-stage sampling plans by attributes, Metrika, 33 (1986), 93–109.
  3. [3] H.W. Kuhn and A.W. Tucker, Nonlinear programming, Proc. of the Second Berkley Symp. on Mathematical Statistics and Probability, University of California Press, Berkley (1951), 481–492.
  4. [4] J.A. Dever, J. Liu, G.V. Iannacchione and D.E. Kendrick, An optimal allocation method for two-stage sampling designs with stratification at the second stage, ASA Proc. of the Joint Statistical Meeting, Amer. Stat. Association, Alexandria, VA (2001).
  5. [5] J. Neyman, On the two different aspects of the representative method: The method of purposive selection, J. of Royal Stat. Soc., 97 (1934), 558–606.
  6. [6] J.P. Ignizio and T.M. Cavalier, Linear Programming, Printice Hall, Englewood Cliff, New Jersey (1994).
  7. [7] J.R. Waters and A.J. Chester, Optimal Allocation in multivariate, twostage sampling designs, The American Statistician, 41 (1987), 46–50.
  8. [8] M.G.M. Khan, E.A. Khan and M.J. Ahasn, An optimal multivariate stratified sampling design using dynamic programming, Australian & New Zealand J. Statist., 45, No 1 (2003), 107–113.
  9. [9] R.G. Clark and D.G. Steel, Optimum allocation of sample to strata and stages with simple additional constraints, J. of the Royal Stat. Soc., Ser. D: The Statistician, 49 (2000), 197–207.
  10. [10] R. Valliant and J.E. Gentle, An application of mathematical programming to sample allocation, Computational Statistics & Data Analysis, 25 (1997), 337–360.
  11. [11] S.-A. Mohammad, The choice of subsample size in two-stage sampling, J. of the Amer. Stat. Association, 81 (1986), 555–558.
  12. [12] W.G. Cochran, Sampling Techniques, Third E d., John Wiley & Sons, Inc., New York (1977).