Comparison Of The Estimation Of Odds Ratio Under Parametric And Non-parametric Methods

Simple random sampling is in common use when we want to get the statistical characteristics of the population with unknown features. However, sometimes we need to consider other sampling methods when it’s hard to sampling or costs lots of time and money by using simple random sampling.In this paper,we will use ranked set sampling and moving extreme ranked set sampling.Fisrtly,we will use numerical modeling to get the estimations of features of the population under simple random sampling,ranked set sampling and moving extreme ranked set sampling.Thus we get the estimation of the distribution of the population.Then we use this estimation to get the odds of the event under study of two population.Hence we get the estimation of odds ratio with respect to specific positive number x0.Secondly,for x0,calculating the estimation of the odds ratio under the method according to Al-Saleh&Samawi (2012).Then comparing the estimation of odds ratio we got under parametric method and non-parametric method.We will show that the bias and mean square error of the estimation of odds ratio under parametric method are smaller.