Your Cart

Your cart is empty

Home  >  Volume 51

14. A MODIFIED ESTIMATOR IN ADAPTIVE CLUSTER SAMPLING by Olayiwola O.M., Ajayi A. O., Aderibigbe T. A., Nurudeen T. S. and Akintola K.A. Volume 51 (May, 2019 Issue)
Sale price: $0.00
Quantity:

A MODIFIED ESTIMATOR IN ADAPTIVE CLUSTER SAMPLING

Olayiwola O.M., Ajayi A. O., Aderibigbe T. A., Nurudeen T. S. and Akintola K.A.

Department of Statistics, College of Physical Sciences, Federal University of Agriculture, Abeokuta, Nigeria

Department of Maths/Statistics, Lagos State Polytechnic, Ikorodu.

Department of Statistics, Oyo State College of Agriculture, Igboora

Abstract

Adaptive cluster sampling is an efficient method for estimating rare and hidden clustered population sizes. Researchers have worked on estimator in adaptive cluster sampling using linear combination of coefficient of skewness, kurtosis and variation. There is dearth of information on estimators based on linear combination of median and coefficient of kurtosis to handle problem of outliers in data sets. This study modified the existing estimator in adaptive cluster sampling using linear combination of median and coefficient of skewness. The data on cotton production (metric tonnes) in Nigeria for the year 2005 and 2006 were extracted from 2013 publications of National Bureau of Statistics. Adaptive cluster sampling was used to select ten cotton producing states (Adamawa, Bauchi, Borno, Gombe, Kaduna, Kano, Katsina, Niger, Sokoto, and Zamfara) at random. Cotton production for 2005 and 2006 were considered as the auxiliary variable and variable of interest respectively. Descriptive statistics and box plot were used to describe the data sets. The large sample properties of the proposed estimator were studied up to the first order of approximation. The bias and Mean Square Error (MSE) for the proposed estimator was obtained and compared with that of the existing estimator to determine its efficiency. Data analysis in r-programming was used to analyse the dataset. Descriptive statistics showed that the average production of cottons for 2005 and 2006 were 46.03 and 48.72 metric tonnes respectively. Box plot showed the presence of outliers in the data set. The median and coefficient of skewness for cotton production in 2005 were 25.11 and 0.55 respectively. The biasness and MSE for the proposed estimator are -0.25 and 0.76 respectively, while for the existing estimator were 0.93 and 3.74 respectively. The modified estimator solved the problems of outliers in data sets efficiently with minimum MSE, hence more efficient.

Close
Loading...