Open navigation menu
Close suggestions
Search
Search
en
Change Language
Upload
Sign in
Sign in
Download free for days
0 ratings
0% found this document useful (0 votes)
42 views
6 pages
SM Lec-2
these are the lecture notes of sampling method
Uploaded by
Åbd Ür Råhmåñ
AI-enhanced title
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content,
claim it here
.
Available Formats
Download as PDF or read online on Scribd
Download
Save
Save SM Lec-2(3) For Later
Share
0%
0% found this document useful, undefined
0%
, undefined
Print
Embed
Report
0 ratings
0% found this document useful (0 votes)
42 views
6 pages
SM Lec-2
these are the lecture notes of sampling method
Uploaded by
Åbd Ür Råhmåñ
AI-enhanced title
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content,
claim it here
.
Available Formats
Download as PDF or read online on Scribd
Carousel Previous
Carousel Next
Download
Save
Save SM Lec-2(3) For Later
Share
0%
0% found this document useful, undefined
0%
, undefined
Print
Embed
Report
Download
Save SM Lec-2(3) For Later
You are on page 1
/ 6
Search
Fullscreen
ESTIMATION is a branch of statistical inference or decision theory. I deals with the estimation of unknown characteristic of « population, IU is defined as “The entire process of using an estimator to produce an estimate of unknown parameter of population is called estimation’ Estimator: Aa statistic used to estimate a parameter is called an estimator, It is also known as a decision function eg the sample mean x is an estimator of j and p isan estimator of \ value obtained trom sample information by applying a rule or formula is called an estimate of population parameter from the s ample was drawn, e.g. the mean height of 20 students taken randomly trom the unive y enrolled students is 65 inches, so 65 inches is an estimate of the averay height of the all students enrolled in the university. Methods of Estimation: There are two methods of estimat 1. The Point Estimation 2. The Interval Estimation 1. The Point-Estimation A process of finding a single value used to estimate @ population paramete is known as point estimation. e.g. The 5 inches is a point estimate of the population mean fi. terval Estimation A process of finding range of values used to estimate a population parameter is known as interval estimation, e.g. av interval estimation of the population mean wis a process of finding an interval like sucb with some level of confidence. Where and “b’ depend upon the point estimator for a particular sample chosen and also on the ampling distribution of the estimator Properties of Goad 1 Estimator A point estimator is said to be a good estimator i t possesses the following properties: 2 Uubiasne; An estimator is said to be unbiased, if the statistic used as an estimator has its expected value equal to the true value of the population parameter being estimated, I 8 is an estimator of the parame! cr 0, then 8 will be called an unbiased estimator iff nd if ‘ ‘Scanned with CamScanneri) (8) 40. then EQ) =e we say that @ isa biased estimator of 0. e.g. ¥ is an unbiased estimator of 41 ats Consisteney An estimator is said to be consistenée, if the statistic used as an estimator becomes closer and closer to the true value of the population parameter being estimated as sample size increases. If @ is an timator of the parameter 0, then 8 will be called a consistent estimator if n — and 8 —> 0. e.g: ¥ isa consistent estimator of wasn % and ¥ > yt {Note: An unbiased estimator is always a consi: -stimator but a consistent estimator may or may ent not be an unbiased} Etfcienes An unbiased estimator is said to be an efficient if the variance of its sampling distribution is less than that of the variance of the sampling distribution of any other unbiased estimator of the same parameter. Suppose there are two unbiased estimators @,and B,of the parameter 0. If var(é,) < var(8.), then 8, will be an efficient estimator. The particular example is that the sample mean (¥ ) and the sample median (%) are two unbiased e imators of the population mean pt, but the var( ¥) < var( 2), so the sample mean (¥ ) is an elf ent estimator as compare to the sample median (2) Sufficien An estimator is said to be a sufficient when no other statistic computed from the same sa ample can provide any additional information oun the parameter being estimated. In other words, sufficiency can be defined as “a statistic used ‘an estimator uses all the information that are contained in the sample, For example the sample mean (¥ ) uses all the into: ation contained in the sample while the sample median (*) does not. - Mean Square Liror (MSE), is defined as the expected value of the squared differences between all possible values of estimator nd the true value of the population parameter. Let é be any estimator (either biased or unbiased), then 1¢ MSE about the true value of 0 is given as MSE@) =~ E(B of = Efé-E(8) + E[6- Ey + = Var(8) + Bias? [ECE) ~ of 2 -14)} 4 E03 ~2 {E(8)~ E(B} 6 ECS) -0 4 = Var(8) + Bias? -2 {0}4 (4) —0 - Var(8) + Bias Variance of statistic minus square of bias If B isan unbinced detimatae dn ton ee ‘Scanned with CamScannerjased estimators when there is no unbiased Use of MSE: It is used to compare the efficieney of two b ; 1 smaller MSE will be used to estimator to estimate the value of parameter, A biased estimator having smaller MSE will estimate the parameter Contidence Interval wn probability (95% or An interval constructed trom the sample data in such a way that it has a known probability : parameter to be estimated, then the 99%) of containing a parameter to be estimated. Suppose 0 is a parameter to be estimated, t ven us: ity that two bounds ‘a’ and *b’ would contain the tue value of 0 i PlasOsb]=1-a tis stated as “we are 100(1 ~ a)% confident that the interval from “a” to “b* contains the tue va probs hue of 0. ain the tue Where « is the level of significance or the probability that the interval {a,b} does not Value of U. The values of a’ and *b’ depend upon the sampling distribution of the statistic or estimator used to estimate 0. In practice the interval estimate is used to estimate a parameter as it is a reliable estimate. Estimation of the population meai yt 1. {he point estnation of ut : The population mean t can be estimated by using statistic ¥ or £ The sampling distribution of the ¥ or # are centered at yt. but the Var(* ) is smaller than Var(.) or any other estimator, Hence in most of the applications ¥ is used as a point estimate for the population mean wt. So we take sample mean ¥ asa point estimate of stimation of u : Let ¥ be the point estimator of 1. The sampling distribution of ¥ is Fou )and the Standard Normal Variate (SNV) is z= ” avn wand +Z.o are the two values in the distribution of SNV which contain the value of Z at (a) point ¥ with probability (1 - a). Hence the 100(1 — u)% confidence interval of Z may be written as (1) 1001 ~ a)% confidence interval of the population mean yt may be obtained by putting the value Z trom (a) in (1). Le. = multiplying each sides by of in = us G/Vn <— fh <-F 42) X0/ Vn subtracting from each sides ¥ > tus o/Vn > W> 24
) at center of each interval indicates the position of the point estimate for each ¥ random sample, Most of the intervals are seen to contain jt but not in every case, All of these intervitls are of the same width, as the width of each interval depends only on the choice of ze. The larger the value of Zu, We chose. the wider we make all intervals and more confident we can be that the particular sample selected will produce an interval that contains the unknown parameter 1 To compute 1001 -af confidence interval for yt, we have assumed that 62 is known. Since this is generally not the ease, we shall repla ice 6 by the sample standard deviation s, provided that n > 30. {sampics A sample of 16 values is selected trom a normal population with unknown mean y and known variance 4. The sample mean was obtained 6.2. find 90%, 92%. 94%, 95% (5.22, 7.18). 96%. 98% and 99% confidence intervals for the true mean of the population. Using the following expression we will obtain 7 different intervals. pecs x vn Solve some more problems, ‘Scanned with CamScanner2: Draw all possible samples of's 1821.2 3 without replacement from the population 3 NN 30 and find mean and median from each sample, Verily that both estisnaly aire unbiased and sample mean is more efficient estimator as evmpare ty the medi Also construct 95% confidence interval for cach sample using sample mean. Coun intervals containing tue mean and not containing true mean, find pereentayes of the intervals and interpret the results. duestion: The mean and standard deviation for quality grade point averages of a random sample of 36 colleges are calculated to be 2.6 and 0.3 respectively. Kind 95%. 99% contidenee inte! the population mean, (2.50, 2.7), (2.471, 2.729) tion 3 inches Question: A sample of 40 observations from a population with standard devi sample mean 64.2 inches. Find 90% confidence: 64.98 inches) iMerval for population mei How bar sa sample size should be for estimating the population’s me is used as un estimate of p. we can be 100(1-0)% confident that th error (e = exceed a speci ed aniount e when the sample size is To estimate the sample size all fractional values are rounded up tw the next whole number - The above formul is applicable if ois known. Ifo is unknown a preliminary sample of size n 2 30 should be se ted to provide an estimate of ©. By using the above formula, we can determine Approximately how many observations are needed to provide the desired degree of accuracy. ‘Scanned with CamScannerIf we draw all possi amples of size n fiom a population, find mean x for cach sample and construct 100(1-a)% vontidence interval for eae ample, these confidence intervals can be shown a Different samples will yield d The ci ular dots (>) at center of each interval indicates ion of the point estimate for each ¥ random sample. Most of the intervals are seen to contain yt but not in every ease. All of these intervals are of the same width, 1s the width of each interval depends only on the choice of z.; The larger the Value of Zu We chose, the wider we make all intervals and more confident we ean be that the particular imple selec ed will produce an interval that contains the unknown parameter 1. ‘To compute 100(1-a)¥ confidence interval for jt, we have assumed that 6° is known, Since this is gen ally not the case. we shall replace 6 by the sample standard deviation S, provided that n> 30 Fsampie: A sample of 16 values is selected from a normal population with unknown mean wand find 90%, 92%, 94%, 7.18). 96%, 98% and 99% confidence intervals for the true mean of the population. Using the following expression we will obtain 7 different intervals. known variance 4. The sample mean was obtained 6, 92 Solve some more problems. ‘Scanned with CamScanner
You might also like
Chapter Two Stat II
PDF
No ratings yet
Chapter Two Stat II
20 pages
Chapter 2 Estimation
PDF
No ratings yet
Chapter 2 Estimation
27 pages
MGMT 222 Ch. IV
PDF
50% (2)
MGMT 222 Ch. IV
30 pages
Theory of Estimation
PDF
No ratings yet
Theory of Estimation
24 pages
Cha 2
PDF
0% (1)
Cha 2
23 pages
Unit 5
PDF
No ratings yet
Unit 5
17 pages
Unit 5 Estimation: Structure
PDF
No ratings yet
Unit 5 Estimation: Structure
17 pages
Inferential PDF
PDF
No ratings yet
Inferential PDF
9 pages
Offiwiz File
PDF
No ratings yet
Offiwiz File
46 pages
Interval Estimation
PDF
No ratings yet
Interval Estimation
69 pages
Lecture 11
PDF
100% (1)
Lecture 11
33 pages
Ch-1.Ppt Business Statx
PDF
No ratings yet
Ch-1.Ppt Business Statx
66 pages
Statistical Inference Point Estimators Estimating The Population Mean Using Confidence Intervals
PDF
No ratings yet
Statistical Inference Point Estimators Estimating The Population Mean Using Confidence Intervals
40 pages
Chapter 5 - Estimation
PDF
No ratings yet
Chapter 5 - Estimation
8 pages
Unit-3 (Estimation)
PDF
No ratings yet
Unit-3 (Estimation)
16 pages
Stat CH 3 Edited 1
PDF
No ratings yet
Stat CH 3 Edited 1
9 pages
Stat For Fin CH 4 PDF
PDF
No ratings yet
Stat For Fin CH 4 PDF
17 pages
ND Vohra Ch10 Theory of Estimation
PDF
No ratings yet
ND Vohra Ch10 Theory of Estimation
37 pages
Lecture 6 Estimation
PDF
No ratings yet
Lecture 6 Estimation
8 pages
Chapter 6
PDF
No ratings yet
Chapter 6
33 pages
Statistics For Economists Lecture VI
PDF
No ratings yet
Statistics For Economists Lecture VI
33 pages
Business Statistics CH 2
PDF
No ratings yet
Business Statistics CH 2
49 pages
Chapter Two
PDF
No ratings yet
Chapter Two
154 pages
Chapter Two (Estimation and Hypothesis Testing)
PDF
No ratings yet
Chapter Two (Estimation and Hypothesis Testing)
20 pages
POINT INTERVAL Estimates
PDF
No ratings yet
POINT INTERVAL Estimates
48 pages
Estimation
PDF
No ratings yet
Estimation
6 pages
Chapter 8
PDF
No ratings yet
Chapter 8
19 pages
Estimation
PDF
No ratings yet
Estimation
53 pages
webMATH236 Lecture5
PDF
No ratings yet
webMATH236 Lecture5
87 pages
Biostat Inferential Statistics
PDF
No ratings yet
Biostat Inferential Statistics
62 pages
11 Parameter Estimation
PDF
No ratings yet
11 Parameter Estimation
101 pages
Module 5
PDF
No ratings yet
Module 5
67 pages
Chapter 2 Statistics Estimation Final
PDF
No ratings yet
Chapter 2 Statistics Estimation Final
13 pages
BS IMI U4 Oct23 Complete
PDF
No ratings yet
BS IMI U4 Oct23 Complete
182 pages
Inferential Statistic: 1 Estimation of A Population Mean
PDF
No ratings yet
Inferential Statistic: 1 Estimation of A Population Mean
8 pages
UMass Stat 516 Solutions Chapter 8
PDF
No ratings yet
UMass Stat 516 Solutions Chapter 8
26 pages
Point and Interval Estimation-26!08!2011
PDF
No ratings yet
Point and Interval Estimation-26!08!2011
28 pages
Chapter Four
PDF
No ratings yet
Chapter Four
9 pages
Introduction To Estimation: OPRE 6301
PDF
100% (1)
Introduction To Estimation: OPRE 6301
18 pages
Ce (PC) 602
PDF
No ratings yet
Ce (PC) 602
21 pages
Chapter 4 - BUSINESS STATISTICS
PDF
No ratings yet
Chapter 4 - BUSINESS STATISTICS
14 pages
CH Ii Business Stat
PDF
No ratings yet
CH Ii Business Stat
28 pages
Estimation
PDF
No ratings yet
Estimation
14 pages
Unit V Estimation
PDF
No ratings yet
Unit V Estimation
33 pages
Flipped Notes 7 Estimation
PDF
No ratings yet
Flipped Notes 7 Estimation
36 pages
Unit 6a Point and Interval Estimation
PDF
No ratings yet
Unit 6a Point and Interval Estimation
13 pages
Learning Objectives
PDF
No ratings yet
Learning Objectives
20 pages
Ch.3-Estimation Module
PDF
No ratings yet
Ch.3-Estimation Module
27 pages
Lecture 4 Estimating From A Sample Sample Mean and Confidence Intervals Upload 2
PDF
No ratings yet
Lecture 4 Estimating From A Sample Sample Mean and Confidence Intervals Upload 2
22 pages
BBA IV Business Statistics
PDF
No ratings yet
BBA IV Business Statistics
270 pages
Lecture 8
PDF
No ratings yet
Lecture 8
85 pages
03 Statistical Inference v0 0 22052022 050030pm
PDF
No ratings yet
03 Statistical Inference v0 0 22052022 050030pm
17 pages
7 Estimation
PDF
No ratings yet
7 Estimation
91 pages
1 Review of Basic Concepts - Interval Estimation
PDF
No ratings yet
1 Review of Basic Concepts - Interval Estimation
4 pages
SSC Gds Notes
PDF
No ratings yet
SSC Gds Notes
88 pages
Estimation
PDF
No ratings yet
Estimation
41 pages
Session: 27: Topic
PDF
No ratings yet
Session: 27: Topic
62 pages
Statistical Estimation
PDF
No ratings yet
Statistical Estimation
39 pages
Stimation: Statistic
PDF
No ratings yet
Stimation: Statistic
46 pages