International Science Community AssociationResearch Journal of Mathematical and Statestical Sciences2320 - 60477220190512An efficient method of generation of a random sample using random numbers significantly less than the sample size2729ENDavid Adugh Kuhe1Department of Mathematics/Statistics/Computer Science, Federal University of Agriculture, Makurdi, Benue State, NigeriaVarun Gangadhar1Department of Agricultural Statistics, Applied Mathematics and Computer Science, Uni. of Agricultural Sciences, Bengaluru, Karnataka, IndiaPooja B.S.2Department of Agricultural Statistics, Applied Mathematics and Computer Science, Uni. of Agricultural Sciences, Bengaluru, Karnataka, IndiaBishvajit Bakshi3Department of Agricultural Statistics, Applied Mathematics and Computer Science, Uni. of Agricultural Sciences, Bengaluru, Karnataka, IndiaPramit Pandit4Department of Agricultural Statistics, Bidhan Chandra KrishiViswavidyalaya, Mohanpur, Nadia, West Bengal, IndiaFalade K.I.1Department of Mathematics, Faculty of Computing and Mathematical Sciences, Kano University of Science and Technology, P.M.B 3244 Wudil Kano State, NigeriaAbubakar A.S.2Department of Mathematics, Faculty of Computing and Mathematical Sciences, Kano University of Science and Technology, P.M.B 3244 Wudil Kano State, NigeriaRushali Gupta1Department of Computer Science and Engineering, BIT Mesra, Ranchi-835215, IndiaSoubhik Chakraborty2Department of Mathematics, BIT Mesra, Ranchi-835215, India>
Short Communication201812820190512Given that the pseudo-random numbers generated by the computer have a cycle; it is wise not to lose random numbers in simulation studies. For drawing a random sample of size n from a population of size N (n<=N), the existing sampling algorithms require n pseudo-random numbers. If N is large, accordingly n should also be large for better representation of the population. Since most simulation studies require at least 500 samples, we would need 500xn pseudo random numbers which can lead to cycle break. We are therefore motivated to develop an efficient sampling algorithm which generates the desired sample using random numbers significantly less than the sample size. Our algorithm has the facility that a single pseudo-random number can generate the sample of size 60 for a population of size 100000 using a python code. We would of course need more than one pseudo-random number if the sample size exceeds 60 for this population.Copyright@ International Science Community Association