An Efficient Class of Estimators for the Finite Population Mean in Ranked Set Sampling

Similar documents
Estimation of Current Population Variance in Two Successive Occasions

Optimal Coordination of Samples in Business Surveys

[Saxena, 2(9): September, 2013] ISSN: Impact Factor: INTERNATIONAL JOURNAL OF ENGINEERING SCIENCES & RESEARCH TECHNOLOGY

An Unbiased Estimator For Population Mean Using An Attribute And An Auxiliary Variable

SIMPLE LINEAR REGRESSION

Chapter 12 Simple Linear Regression

Suggested Answers To Exercises. estimates variability in a sampling distribution of random means. About 68% of means fall

Comparing Means: t-tests for Two Independent Samples

μ + = σ = D 4 σ = D 3 σ = σ = All units in parts (a) and (b) are in V. (1) x chart: Center = μ = 0.75 UCL =

Stratified Analysis of Probabilities of Causation

Numerical algorithm for the analysis of linear and nonlinear microstructure fibres

Stochastic Optimization with Inequality Constraints Using Simultaneous Perturbations and Penalty Functions

SAMPLING. Sampling is the acquisition of a continuous signal at discrete time intervals and is a fundamental concept in real-time signal processing.

Bio 112 Lecture Notes; Scientific Method

Alternate Dispersion Measures in Replicated Factorial Experiments

Lecture 7: Testing Distributions

Source slideplayer.com/fundamentals of Analytical Chemistry, F.J. Holler, S.R.Crouch. Chapter 6: Random Errors in Chemical Analysis

ON THE APPROXIMATION ERROR IN HIGH DIMENSIONAL MODEL REPRESENTATION. Xiaoqun Wang

1. The F-test for Equality of Two Variances

Clustering Methods without Given Number of Clusters

Social Studies 201 Notes for November 14, 2003

Beta Burr XII OR Five Parameter Beta Lomax Distribution: Remarks and Characterizations

Z a>2 s 1n = X L - m. X L = m + Z a>2 s 1n X L = The decision rule for this one-tail test is

Preemptive scheduling on a small number of hierarchical machines

Multipurpose Small Area Estimation

2 Model-assisted and calibration estimators for finite population totals

A Bluffer s Guide to... Sphericity

By Xiaoquan Wen and Matthew Stephens University of Michigan and University of Chicago

CHAPTER 8 OBSERVER BASED REDUCED ORDER CONTROLLER DESIGN FOR LARGE SCALE LINEAR DISCRETE-TIME CONTROL SYSTEMS

Social Studies 201 Notes for March 18, 2005

Lecture 4 Topic 3: General linear models (GLMs), the fundamentals of the analysis of variance (ANOVA), and completely randomized designs (CRDs)

7.2 INVERSE TRANSFORMS AND TRANSFORMS OF DERIVATIVES 281

Asymptotics of ABC. Paul Fearnhead 1, Correspondence: Abstract

A Class of Linearly Implicit Numerical Methods for Solving Stiff Ordinary Differential Equations

If Y is normally Distributed, then and 2 Y Y 10. σ σ

CHAPTER 4 DESIGN OF STATE FEEDBACK CONTROLLERS AND STATE OBSERVERS USING REDUCED ORDER MODEL

ARTICLE Overcoming the Winner s Curse: Estimating Penetrance Parameters from Case-Control Data

CHAPTER 6. Estimation

Efficient Methods of Doppler Processing for Coexisting Land and Weather Clutter

UNCLAS I FW D UNC LAS,1 1F] ED CENTER DEFENSE DOCUIVIENIAI','TAON. TECHNICAL INfFORMATION SCIENTIFIC AND. CAMEHON STATION, ALEXANDRIA, MIlIGINIA

Finite Element Analysis of a Fiber Bragg Grating Accelerometer for Performance Optimization

MINITAB Stat Lab 3

GNSS Solutions: What is the carrier phase measurement? How is it generated in GNSS receivers? Simply put, the carrier phase

Lecture 21. The Lovasz splitting-off lemma Topics in Combinatorial Optimization April 29th, 2004

Small Area Estimation Under Transformation To Linearity

Standard Guide for Conducting Ruggedness Tests 1

Extension of Inagaki General Weighted Operators. and. A New Fusion Rule Class of Proportional Redistribution of Intersection Masses

Chapter 4. The Laplace Transform Method

Asymptotic Values and Expansions for the Correlation Between Different Measures of Spread. Anirban DasGupta. Purdue University, West Lafayette, IN

One Class of Splitting Iterative Schemes

Appendix. Proof of relation (3) for α 0.05.

Research Article A New Kind of Weak Solution of Non-Newtonian Fluid Equation

Comparison of Estimators in Case of Low Correlation in Adaptive Cluster Sampling. Muhammad Shahzad Chaudhry 1 and Muhammad Hanif 2

New bounds for Morse clusters

Suggestions - Problem Set (a) Show the discriminant condition (1) takes the form. ln ln, # # R R

Week 3 Statistics for bioinformatics and escience

A NEW OPTIMAL QUADRATIC PREDICTOR OF A RESIDUAL LINEAR MODEL IN A FINITE POPULATION

Bogoliubov Transformation in Classical Mechanics

Pairwise Markov Random Fields and its Application in Textured Images Segmentation

Design By Emulation (Indirect Method)

Chapter 2 Sampling and Quantization. In order to investigate sampling and quantization, the difference between analog

Simple Observer Based Synchronization of Lorenz System with Parametric Uncertainty

A Study on Simulating Convolutional Codes and Turbo Codes

Dynamical Behavior Analysis and Control of a Fractional-order Discretized Tumor Model

arxiv: v1 [cond-mat.stat-mech] 22 Sep 2009

Lecture 3. January 9, 2018

Predicting the Shear Capacity of Reinforced Concrete Slabs subjected to Concentrated Loads close to Supports with the Modified Bond Model

ASSESSING EXPECTED ACCURACY OF PROBE VEHICLE TRAVEL TIME REPORTS

LTV System Modelling

Reliability Analysis of Embedded System with Different Modes of Failure Emphasizing Reboot Delay

Lecture 10 Filtering: Applied Concepts

Solutions to Supplementary Problems

A Simplified Methodology for the Synthesis of Adaptive Flight Control Systems

Research Article Reliability of Foundation Pile Based on Settlement and a Parameter Sensitivity Analysis

MATEMATIK Datum: Tid: eftermiddag. A.Heintz Telefonvakt: Anders Martinsson Tel.:

A new approach to determinate parasitic elements of GaN HEMT by COLD FET S-Parameter

EC381/MN308 Probability and Some Statistics. Lecture 7 - Outline. Chapter Cumulative Distribution Function (CDF) Continuous Random Variables

Thermal Resistance Measurements and Thermal Transient Analysis of Power Chip Slug-Up and Slug-Down Mounted on HDI Substrate

Parametrization of the 511 kev respond in BGO/ LSO Crystals with respect to Spatial Resolution in PETR/CT Scans

Convex Hulls of Curves Sam Burton

A Simplified Dynamics Block Diagram for a Four-Axis Stabilized Platform

Theoretical Computer Science. Optimal algorithms for online scheduling with bounded rearrangement at the end

Feasible set in a discrete epidemic model

Non-linearity parameter B=A of binary liquid mixtures at elevated pressures

Unavoidable Cycles in Polynomial-Based Time-Invariant LDPC Convolutional Codes

( ) ( Statistical Equivalence Testing

Statistics and Data Analysis

Stochastic Neoclassical Growth Model

time? How will changes in vertical drop of the course affect race time? How will changes in the distance between turns affect race time?

Gain and Phase Margins Based Delay Dependent Stability Analysis of Two- Area LFC System with Communication Delays

SIMPLIFIED MODEL FOR EPICYCLIC GEAR INERTIAL CHARACTERISTICS

Jan Purczyński, Kamila Bednarz-Okrzyńska Estimation of the shape parameter of GED distribution for a small sample size

Efficiency Analysis of a Multisectoral Economic System

Combining allele frequency uncertainty and population substructure corrections in forensic DNA calculations

ON A CERTAIN FAMILY OF QUARTIC THUE EQUATIONS WITH THREE PARAMETERS. Volker Ziegler Technische Universität Graz, Austria

A Method for Assessing Customer Harmonic Emission Level Based on the Iterative Algorithm for Least Square Estimation *

Reformulation of Block Implicit Linear Multistep Method into Runge Kutta Type Method for Initial Value Problem

NULL HELIX AND k-type NULL SLANT HELICES IN E 4 1

Remote Estimation of Correlated Sources under Energy Harvesting Constraints

Problem Set 8 Solutions

Transcription:

Open Journal of Statitic 06 6 46-435 Publihed Online June 06 in SciRe http://cirporg/journal/oj http://ddoiorg/0436/oj0663038 An Efficient Cla of Etimator for the Finite Population Mean in Ranked Set Sampling Lakhkar Khan Javid Shabbir Department of Statitic Government College Toru Khber Pukhtunkha Pakitan Department of Statitic Quaid-i-Azam Univerit Ilamabad Pakitan Received 4 Februar 06; accepted June 06; publihed 4 June 06 Copright 06 b author Scientific Reearch Publihing Inc Thi ork i licened under the Creative Common Attribution International Licene (CC BY http://creativecommonorg/licene/b/40/ Abtract In thi paper e propoe a cla of etimator for etimating the finite population mean of the tud variable under Ranked Set Sampling ( hen population mean of the auiliar variable i knon The bia Mean Squared Error (MSE of the propoed cla of etimator are obtained to firt degree of approimation It i identified that the propoed cla of etimator i more efficient a compared to [] etimator everal other etimator A imulation tud i carried out to judge the performance of the etimator Keord Ranked Set Sampling Auiliar Variable Bia Mean Squared Error Relative Efficienc Introduction The problem of etimation in the finite population mean ha been idel conidered b man author in different ampling deign In application there ma be a ituation hen the variable of interet cannot be meaured eail or i ver epenive to do o but it can be ranked eail at no cot or at ver little cot In vie of thi ituation [] introduced the Ranked Set Sampling ( procedure [3] proved the mathematical theor that the ample mean under a an unbiaed etimator of the finite population mean more precie than the ample mean etimator under imple rom ampling (SRS The auiliar information pla an important role in increaing efficienc of the etimator [4] uggeted an etimator for population ratio in hoed that it had le variance a compared to uual ratio etimator in imple rom ampling (SRS In perfect ranking of element a conidered b [] [3] for etimation of population mean In ome ituation ranking ma not be perfect According to [5] the ample mean in i an unbiaed etimator of the Ho to cite thi paper: Khan L Shabbir J (06 An Efficient Cla of Etimator for the Finite Population Mean in Ranked Set Sampling Open Journal of Statitic 6 46-435 http://ddoiorg/0436/oj0663038

L Khan J Shabbir population mean regardle of error in ranking of the element In [6] the ranking of element a done on bai of the auiliar variable intead of judgment [] uggeted an etimator for population mean ranking of the element a oberved on bai of the auiliar variable [7] had uggeted a cla of Hartle-Ro tpe unbiaed etimator in [8] had alo propoed unbiaed etimator in tratified ranked et ampling In thi paper e ugget a cla of etimator for the population mean uing knon population mean of the auiliar variable in It i hon that the propoed cla of etimator outperform a compared to the [9] [] everal other etimator Alo ome pecial cae of the propoed cla are conidered in Table A (Appendi Ranked Set Sampling Procedure In ranked et ampling ( e elect m rom ample each of ize m unit from the population rank the unit ithin each ample ith repect to a variable of interet In order to facilitate the ranking the deign parameter m i choen to be mall From the firt ample the unit having the loet rank i elected from the econd ample the unit having econd loet rank i elected the proce i continued until from the lat ample the unit having the highet rank i elected In thi a e obtain m meaured unit one from each ample The ccle ma be repeated r time until mr unit have been meaured Thee n mr unit form the data Suppoe that the variable of interet Y i difficult to meaure to rank but there i the auiliar variable X hich i correlated ith Y The variable X ma be ued to obtain the rank of Y To perform the ampling procedure m bivariate rom ample each of ize m unit are dran from the population then each ample i ranked ith repect to one of the variable Y or X Here e aume that the perfect ranking i done on bai of the auiliar variable X hile the ranking of Y i ith error An actual meaurement from the firt ample i then taken of the unit ith the mallet rank of X together ith the variable Y aociated ith the mallet rank of X From the econd ample of ize m the Y aociated ith the econd mallet rank of X i meaured The proce i continued until from the mth ample the Y aociated ith the highet rank of X i meaured The ccle i repeated r time until n mr bivariate unit have been meaured out of the total mr elected unit 3 Some Eiting Etimator Notation We conider a ituation hen rank the element on the auiliar variable Let [ ] i j ( i be the ith judgment j ordering in the ith et for the tud variable Y baed on the ith order tatitic of the ith et of the auiliar variable X at the jth ccle Baed on the ample mean etimator ( of the population mean ( Y i given b [ ] ( here [ ] ( mr j i [ i] j r m To obtain the bia MSE of etimator e define: uch that here Y e X e [ ] ( 0 ( E( e E e ( 0 γ E e C W 0 0 E e γ C W E ee γρcc W 0 m m m W τ W W [ ] i τ i τ i m rxy m rx m ry i i i ( τ µ i i X τ [ ] µ i i [ ] Y τ µ i i [ ] Y µ X C i ρcc C C are the 47

L Khan J Shabbir coefficient of variation of Y X repectivel It alo be noted that the value of µ i [ ] µ i are the mean of ith order tatitic from ome pecific ditribution (ee [0] The variance of under cheme i given b [4] propoed an etimator of the population ratio ( ( γ Var Y C W ( Y R under a: X ˆ [ ] R (3 ( When population mean ( X of the auiliar variable (X i knon the variable Y X are poitivel correlated [9] propoed the ratio etimator for population mean (Y baed on a The bia MSE of r r [ ] X (4 ( up to the firt degree of approimation are given b ( r γ ( ρ ( Bia Y C C C W W ( r γ ( ρ ( MSE Y C C C C W W W When population mean ( X of the auiliar variable (X i knon the variable Y X are negativel correlated then the product etimator baed on i defined a: The bia MSE of p ( [ ] X (7 p up to the firt degree of approimation are given b ( p ( γρ Bia Y C C W (8 ( p { γ ( ρ ( } MSE Y C C C C W W W (9 [] uggeted an etimator under i defined a: here λ i uitabl choen contant The minimum bia MSE of at optimum value of λ ie are given b λ Bia ( opt MSE λ [ ] (0 ( γ C W min min Y ( γ C W ( γ C W Y ( γ C W ( γ C W The difference-tpe etimator for population mean ( Y baed on i given b (5 (6 ( ( 48

L Khan J Shabbir [ ] d d ( X ( here d i a contant The minimum variance of d( at optimum value of d ie i given b d ( opt (3 ( γ ( γ C W R C W Var ( d Y γ C min W ( γ C W ( γ C W Folloing [] [] uggeted a cla of etimator of the population mean (Y baed on a: g ax b λ[ ] λ [ ] S α( a b ( α ( ax b here α i a uitabl choen contant a b are either real number or function of knon parameter of the auiliar variable X g i a calar hich take value of (for generating ratio-tpe etimator (for generating product-tpe etimator ( λ λ are contant hoe um need not be unit The bia of S( i given b g ( S λ λ λ α θ ( γ λ αθ γρ Y Bia Y ( g C W g C C W The MSE of S( to firt degree of approimation i given b here ( λ λ S λλ λ λ MSE Y A A B B C C D D ( γ C A A W {( } B γ C g g θ α C 4 gαθc B W g g θ α W 4 gαθw ( g g C C g C C γ θα θ α ( g g C W gθαc θ α W ( g g D γ θ α C gθαc ( We dicu to cae Cae : Sum of eight i unit (ie λ λ Solving (7 the optimum value of λ i given b g g D θ α W gθαw (4 (5 (6 (7 49

L Khan J Shabbir ( B B ( C C ( D D ( A A ( B B ( C C λ opt λ in (7 e get the minimum MSE of S( given b ( B B C C D D MSE ( Y ( ( S B B D D min ( A A ( B B ( C C Subtituting ( opt Cae : Sum of eight i fleible (ie λ λ Solving (7 the optimum value of λ λ are given b ( B B ( C C ( D D ( opt ( A A( B B ( C C λ λ ( A A ( D D ( C C ( opt ( A A( B B ( C C Subtituting the optimum value of λ λ in (7 e get MSE {( B B ( C C ( D D ( A A ( D D } { } ( Y S min ( A A( B B ( C C (8 (9 4 Propoed Cla of Etimator Folloing [] [] e propoe a cla of etimator of the population mean (Y under a ax b a b ax b k[ ] k( X α ep α L ( ax b ( a ( b ( a ( b here α i a uitabl choen contant a b are either real number or the function of knon parameter of the auiliar variable X ( k k are contant hoe um need not be unit From (0 e can generate a large number of etimator for the different value of the contant (Table A in Appendi The propoed etimator L( can be ritten in term of e 0 e a e 3 e ky ( e0 kxe θ θ α ( α( θe L ( 8 here θ ax ( ax b Solving ( e have α 5α Y Y( k ky θe kyθ e kye L 8 0 α α ky θee 0 kxe kx θe Taking epectation of both ide of above equation e get bia of L( given b 5α α Bia Y k kyθ L γc W kyθ C W 8 γ α kx θ ( γc W ( ( (0 ( (3 430

L Khan J Shabbir Squaring both ide of Equation ( ignoring higher order term of e e have α α ( Y Y ( k ky e0 θ e θee L 0 5α α kx e k( k Y θ e θe 8 α α k( k YX θe e kk YX θe e0e Taking epectation of both ide of above equation e obtain the MSE of L( a given b here ( ( ( L MSE Y k k E E k F F k k G G We dicu to cae Cae : Sum of eight i unit (ie k k The optimum value of k i given b k ( opt k k H H kk I I E Y C α α γ θ C θc α α E Y W θ W θw F X C F γ XW 5 G Y α α γ θ C θc 8 5α α G Y θ W θw 8 α H XY γ θc α H XY θ W α I XY C C γ θ α I XY θ W W i Y ( F F ( G G ( H H ( I I Y E E F F G G H H I I Thu the minimum MSE of L( i given b MSE { } { } E E Y H H F F I I G G ( L( min Y ( E E ( F F ( G G ( H H ( I I (4 (5 43

L Khan J Shabbir Cae : Sum of eight i fleible (ie k k For ( k k the MSE of L( in Equation (4 i minimized for { } { } { } { } F F Y G G H H H H I I opt ( F F Y ( G G ( E E H H I I k k { } { } ( H H {( E E ( G G } ( I I Y ( G G { } opt ( F F Y ( G G ( E E H H I I Subtituting the optimum value of k k in (4 e get ( L opt opt opt min MSE Y k k E E k F F ( ( opt ( opt ( k k G G k k H H ( opt ( opt opt opt k k I I Note: It i difficult to make the theoretical comparion due to compleit therefore e adopt the numerical tud 5 Simulation Stud We ue the ame data et a earlier ued b [] perform ome imulation tud to invetigate the performance of the etimator Population (ource: [3] Y Number of acre devoted to farm during 99 (ACRES9 X Number of large farm during 99 (LARGEF9 N 3059 ρ 067748 Y 308584 X 565 S 4538 S 73 We et r 0 m 5 to elect a ample of n mr 50 unit from the population of ize N 3059 To compute the value of W W W b imulation e eplain our imulation methodolog a follo Here W W W can be ritten a m W ( [ ] RDY i mr here W i m mr i mr i ( RDX ( i m W RDX i RDY i [ ] ( ( [ ] µ µ i [ ] i RDY i RDX ( i i m Y X To find the poible value of the ratio RDY [ i ] for m 5 e generate i ~ ( 0 [ ] 05 008e RDY [ ] 050 008e RDY [ 3] 00 008e3 RDY [ 4] 5 008e4 [ 5] 75 008e5 e N calculate RDY RDY It mean that hen the firt mallet value i elected from the ranked et ample the epected ratio of that value to the population mean could be cloe to 05 hen the econd mallet value i (6 43

L Khan J Shabbir elected the ratio of that value to the population mean could be cloe to 050 hen the third mallet value i elected the epected ratio of that value to the population mean ill cloe to Similarl the epected ratio of the fourth fifth value could be cloe to 5 75 repectivel In each cae e eighted error term e i ith a mall number 008 to make ure that the ratio RDY [ i ] remain poitive In other ord it mean that e are generating ei ~ N ( 0 008 Thu the poible value of the ratio RDY [ i ] are epected to remain cloe to thoe e are conidering here Similarl for the poible value of the ratio RDX ( i e conider RDX ( 05 005e RDX 050 005e RDX ( 3 00 005e3 RDX ( 4 5 005e4 RDX ( 5 75 005e5 here ei ~ N ( 0 Here e eighted e i ith a mall number 005 becaue it ma be le rik to rank the auiliar variable X than the tud variable Y Thu the value of W i [ ] W i W are obtained through thi imulation are repreented in the lat three column of Table ( i Table PRE of propoed cla of etimator through imulation a b g α R ( 0 R ( 0 R ( 03 R ( 0 4 R ( 05 ( 06 R W W W i i ( i 5 5 0 406 03 609 64 53 645 000573 000574 000573 5 5 05 393 03 599 605 638 644 000590 000604 000596 5 5 09 48 034 67 675 655 78 00046 000404 00043 5 5 0 445 033 64 648 573 678 00056 000485 000499 5 5 05 34 030 545 568 575 587 000689 000764 00075 5 5 09 446 033 64 686 634 688 00054 00048 000497 5 0 0 369 03 580 586 480 65 00065 000658 00064 5 0 05 376 03 586 59 60 630 00065 00064 00068 5 0 09 45 033 64 69 67 670 000546 000530 000538 5 0 0 30 030 58 537 40 56 00050 00086 000766 5 0 05 409 03 6 635 649 657 000568 000567 000567 5 0 09 37 03 583 67 595 69 00060 00065 000635 5 5 0 408 03 6 66 505 648 000569 000570 000569 5 5 05 403 03 607 6 64 653 000576 00058 000578 5 5 09 35 03 567 573 587 6 000649 000697 000673 5 5 0 38 03 59 599 478 67 000605 00069 00066 5 5 05 39 03 598 6 630 643 00059 00060 000598 5 5 09 433 033 63 680 639 68 000533 00053 0005 5 5 0 334 03 554 560 49 588 00067 000743 000706 5 5 05 408 03 6 66 645 657 000569 000578 000576 5 5 09 403 03 608 63 60 654 000575 000578 000576 5 5 0 43 03 64 63 5 66 000546 000540 00054 5 5 05 453 033 647 67 687 695 000504 000467 000484 5 5 09 39 03 599 643 6 646 00059 000605 000598 5 0 0 334 030 554 560 444 588 00067 000743 000658 5 0 05 408 03 6 66 648 656 000569 000568 000566 5 0 09 459 033 65 656 648 696 000496 000453 000473 5 0 0 43 033 64 63 536 66 000545 000540 000540 5 0 05 46 03 68 64 656 663 000557 00055 000553 5 0 09 403 03 607 65 64 653 000576 00058 000578 5 5 0 39 03 599 604 58 635 00059 000605 000597 5 5 05 330 030 55 557 58 59 000679 000749 00073 5 5 09 373 03 584 589 590 67 00069 000650 000634 5 5 0 47 03 69 64 544 655 000555 00055 00050 5 5 05 43 033 63 646 664 668 000548 000534 000540 5 5 09 35 03 568 60 578 6 000648 00070 00067 433

L Khan J Shabbir We invetigate the percentage relative efficienc (PRE of ratio etimator ˆ r θ (a the Searl etimator ˆ θ the difference etimator ˆ d θ3 [] etimator ˆ S( θ4 hen λ λ ith repect to conventional etimator ˆ θ0 (a We alo calculate PRE of the propoed cla of etimator a ˆ θ L 5 hen ( k k hen ( k k a ˆ θ L 6 ith repect to ˆ θ0 The PRE of our propoed etimator other eiting etimator ˆj θ j 6 ith repect to conventional etimator ˆ θ0 i defined a PRE ( ˆ θ ˆ 0 θ j ( ˆ θ0 ( ˆ θ j MSE MSE 00 j 6 (7 The PRE of our propoed etimator other eiting etimator ith repect to conventional etimator are given in Table 6 Concluion Since abg α are the fied contant in [] etimator in the propoed cla of etimator There can be a large number of combination for different value of thee contant Here onl limited number of reult are reported in Table Obvioul it can be oberved through the imulation tud in Table that the propoed cla of etimator i more efficient than all conidered etimator It PRE increae from 645 to 78 hen α change from 0 to 09 but decreae lightl hen α i cloe to 05 Generall e can a PRE of propoed cla increae a value of α increae for fied value of contant a b g [] Cla of etimator ha maimum PRE 675 but it i le efficient a compared to the propoed cla of etimator for all the choice of contant reported in Table Alo from the Table e can ee that other competitor etimator are alo le efficient than the propoed cla of etimator If e make comparion beteen the to k k i more precie than the Cae propoed cae then the cla of etimator in Cae ( ( k k We can ee from Table that b fiing the value of a b at 5 the propoed clae of etimator give more precie reult hen the value of α i aa form 05 either cloe to 0 or While b fiing poitive value of the contant a b e get more precie reult for α cloe to 05 Therefore the propoed cla of etimator can be preferred over it competitive etimator in application under Acknoledgement The author ih to thank the editor the anonmou referee for their uggetion hich led to improvement in the earlier verion of the manucript Reference [] Singh HP Tailor R Singh S (04 General Procedure for Etimating the Population Mean Uing Ranked Set Sampling Journal of Statitical Computation Simulation 84 93-945 http://ddoiorg/0080/009496550733395 [] Mclntre G (95 A Method for Unbiaed Selective Sampling Uing Ranked Set Crop Pature Science 3 385-390 http://ddoiorg/007/ar950385 [3] Takahai K Wakimoto K (968 On Unbiaed Etimate of the Population Mean Baed on the Sample Stratified b Mean of Ordering Annal of the Intitute of Statitical Mathematic 0-3 http://ddoiorg/0007/bf096 [4] Samai HM Muttlak MA (996 Etimation of Ratio Uing Ranked Set Sampling Biometrical Journal 38 753-764 http://ddoiorg/000/bimj47038066 [5] Dell T Clutter J (97 Ranked Set Sampling Theor ith Order Statitic Background Biometric 545-555 http://ddoiorg/0307/55666 [6] Stoke SL (977 Ranked Set Sampling ith Concomitant Variable Communication in Statitic: Theor Method 6 07- http://ddoiorg/0080/03609770887563 [7] Khan L Shabbir J (05 A Cla of Hartle-Ro Tpe Unbiaed Etimator for Population Mean Uing Ranked Set Sampling Hacettepe Journal of Mathematic Statitic http://ddoiorg/0567/hjms0560579 434

L Khan J Shabbir [8] Khan L Shabbir J (06 Hartle-Ro Tpe Unbiaed Etimator Uing Ranked Set Sampling Stratified Ranked Set Sampling North Carolina Journal of Mathematic Statitic 0- [9] Kadilar C Unazici Y Cingi H (009 Ratio Etimator for the Population Mean Uing Ranked Set Sampling Statitical Paper 50 30-309 http://ddoiorg/0007/0036-007-0079- [0] Arnold BC Balakrihnan N Nagaraja HN (0 A Firt Coure in Order Statitic Vol 54 Siam [] Searl DT (964 The Utilization of a Knon Coefficient of Variation in the Etimation Procedure Journal of the American Statitical Aociation 59 5-6 http://ddoiorg/0080/064599640480765 [] Khohnevian M Singh R Chauhan P Saan N Smarache F (007 A General Famil of Etimator for Etimating Population Mean Uing Knon Value of Some Population Parameter( Far Eat Journal of Theoretical Statitic 8-9 [3] Lohr S (999 Sampling: Deign Anali Dubur Pre Boton Appendi Table A Some pecial cae of the propoed cla of etimator k k α a b Etimator Remark 0 0 0 ( [ ] Uual mean etimator X 0 0 0 r( [ ] ( X λ 0 0 0 λ r( [ ] ( ( β 0 0 reg( [ ] β X ( k k 0 0 d( [ ] k X X k 0 0 [ ] k dr ( X ( ( X k 0 0 k gdr( [ ] k ( X( ( X β 0 0 regr( [ ] β ( X ( ( X 0 0 [ ] ep e X X ( β 0 [ ] β ( X ep rege X ( ( ( Uual ratio etimaotr Kadilar et al (009 ratio tpe etimator Regreion tpe etimator Difference tpe etimator Difference-ratio etimator Generalied difference-ratio etimator Regreion-ratio etimator Eponential tpe etimator Regreion-eponential tpe etimator 435