# properties of an estimator

The closer the expected value of the point estimator is to the value of the parameter being estimated, the less bias it has. Rigorous derivations of the statistical properties of the estimator are provided in the books by Fleming & Harrington  and Andersen et al. The following are desirable properties for statistics that estimate population parameters: Unbiased: on average the estimate should be equal to the population parameter, i.e. If you continue browsing the site, you agree to the use of cookies on this website. Hence an estimator is a r.v. A consistent estimator is an estimator whose probability of being close to the parameter increases as the sample size increases. Let β’j(N) denote an estimator of βj­ where N represents the sample size. Consider the linear regression model where the outputs are denoted by , the associated vectors of inputs are denoted by , the vector of regression coefficients is denoted by and are unobservable error terms. An estimator ^ n is consistent if it converges to in a suitable sense as n!1. Estimator is Best; So an estimator is called BLUE when it includes best linear and unbiased property. 1 (4.6) These results are summarized below. Otherwise, a non-zero difference indicates bias. Properties of O.L.S. BLUE: An estimator is BLUE when it has three properties : Estimator is Linear. 1. It produces a single value while the latter produces a range of values. Point estimation is the opposite of interval estimation. ECONOMICS 351* -- NOTE 4 M.G. 9 Properties of point estimators and nding them 9.1 Introduction We consider several properties of estimators in this chapter, in particular e ciency, consistency and su cient statistics. Putting this in standard mathematical notation, an estimator is unbiased if: There are three desirable properties every good estimator should possess. The bias is the difference between the expected value of the estimator and the true value of the parameter. Measures of Central Tendency, Variability, Introduction to Sampling Distributions, Sampling Distribution of the Mean, Introduction to Estimation, Degrees of Freedom Learning Objectives. There are four main properties associated with a "good" estimator. Suppose we have two unbiased estimators – β’j1 and β’j2 – of the population parameter βj: We say that β’j1 is more efficient relative to β’j2  if the variance of the sample distribution of β’j1 is less than that of β’j2  for all finite sample sizes. 2.2 Finite Sample Properties The first property deals with the mean location of the distribution of the estimator. This allows us to use the Weak Law of Large Numbers and the Central Limit Theorem to establish the limiting distribution of the OLS estimator. If you continue browsing the site, you agree to the use of cookies on this website. It is unbiased 3. Then an "estimator" is a function that maps the sample space to a set of sample estimates. Minimum Variance S3. t is an unbiased estimator of the population parameter Ï provided E[t] = Ï. A point estimator (PE) is a sample statistic used to estimate an unknown population parameter. It uses sample data when calculating a single statistic that will be the best estimate of the unknown parameter of the population. An estimator that has the minimum variance but is biased is not good 11 Characteristics of Estimators. We assume to observe a sample of realizations, so that the vector of all outputs is an vector, the design matrixis an matrix, and the vector of error termsis an vector. The most fundamental desirable small-sample propertiesof an estimator are: S1. An estimator is said to be unbiased if its expected value is identical with the population parameter being estimated. Slideshare uses cookies to improve functionality and performance, and to provide you with relevant advertising. All Rights ReservedCFA Institute does not endorse, promote or warrant the accuracy or quality of AnalystPrep. We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. A good estimator, as common sense dictates, is close to the parameter being estimated. Unbiasedness, Efficiency, Sufficiency, Consistency and Minimum Variance Unbiased Estimator. It is linear (Regression model) 2. Thus, this difference is, and should be zero, if an estimator is unbiased. Now customize the name of a clipboard to store your clips. An estimator Î¸Ë= t(x) is said to be unbiased for a function Î¸ if it equals Î¸ in expectation: E Î¸{t(X)} = E{Î¸Ë} = Î¸. An estimator's expected value (the mean of its sampling distribution) equals the parameter it is intended to estimate. Looks like you’ve clipped this slide to already. Prerequisites. estimator for one or more parameters of a statistical model. ©AnalystPrep. The bias (B) of a point estimator (U) is defined as the expected value (E) of a point estimator minus the value of the parameter being estimated (Î¸). In statistics, "bias" is an objective property of an estimator. The two main types of estimators in statistics are point estimators and interval estimators. Identify and describe desirable properties of an estimator. When some or all of the above assumptions are satis ed, the O.L.S. This intuitively means that if a PE  is consistent, its distribution becomes more and more concentrated around the real value of the population parameter involved. Unbiasedness. This property is more concerned with the estimator rather than the original equation that is being estimated. Note that OLS estimators are linear only with respect to the dependent variable and not necessarily with respect to the independent variables. The property of unbiasedness (for an estimator of theta) is defined by (I.VI-1) where the biasvector delta can be written as (I.VI-2) and the precision vector as (I.VI-3) which is a positive definite symmetric K by K matrix. sample from a population with mean and standard deviation Ë. Some simulation results are presented in Section 6 and finally we draw conclusions in Section 7. A point estimator is a statistic used to estimate the value of an unknown parameter of a population. You can change your ad preferences anytime. It should be unbiased: it should not overestimate or underestimate the true value of the parameter. A biased estimator can be less or more than the true parameter, giving rise to both positive and negative biases. The expected value of that estimator should be equal to the parameter being estimated. Bias is a distinct concept from consistency. It is a random variable and therefore varies from sample to sample. Maximum Likelihood Estimation (MLE) is a widely used statistical estimation method. We define three main desirable properties for point estimators. These properties are defined below, along with comments and criticisms. Note that not every property requires all of the above assumptions to be ful lled. Properties of estimators Unbiased estimators: Let ^ be an estimator of a parameter . See our User Agreement and Privacy Policy. The bias of an estimator Î¸Ë= t(X) of Î¸ is bias(Î¸Ë) = E{t(X)âÎ¸}. CFA® and Chartered Financial Analyst® are registered trademarks owned by CFA Institute. Estimator is Unbiased. Putting this in standard mathematical notation, an estimator is unbiased if: E(β’j) = βj­   as long as the sample size n is finite. 1. It is an efficient estimator (unbiased estimator with least variance) We say that ^ is an unbiased estimator of if E( ^) = Examples: Let X 1;X 2; ;X nbe an i.i.d. Recall: the moment of a random variable is The corresponding sample moment is The estimator based on the method of moments will be the solution to the equation . This video elaborates what properties we look for in a reasonable estimator in econometrics. It is one of the oldest methods for deriving point estimators. Another asymptotic property is called consistency. Linear Estimator : An estimator is called linear when its sample observations are linear function. Four estimators are presented as examples to compare and determine if there is a "best" estimator. Probability is a measure of the likelihood that something will happen. Abbott 2. Properties of the O.L.S. It’s also important to note that the property of efficiency only applies in the presence of unbiasedness since we only consider the variances of unbiased estimators. An estimator of  is usually denoted by the symbol . The OLS estimator is one that has a minimum variance. Unbiasedness S2. The eciency of V â¦ In general, you want the bias to be as low as possible for a good point estimator. See our Privacy Policy and User Agreement for details. In short, if we have two unbiased estimators, we prefer the estimator with a smaller variance because this means it’s more precise in statistical terms. There are three desirable properties of estimators: unbiasedness. In assumption A1, the focus was that the linear regression should be âlinear in parameters.â However, the linear property of OLS estimator means that OLS belongs to that class of estimators, which are linear in Y, the dependent variable. For Example then . An estimator ^ for We could say that as N increases, the probability that the estimator ‘closes in’ on the actual value of the parameter approaches 1. Author(s) David M. Lane. If bias(Î¸Ë) is of the form cÎ¸, Î¸Ë= Î¸/Ë (1+c) is unbiased for Î¸. The first property deals with the population parameter Ï provided E [ t =... Varies from sample to sample estimator with least variance ) there are three desirable properties every estimator! Three properties: 1 the most fundamental desirable small-sample propertiesof an estimator of βj­ where n represents the sample increases... Of being close to the parameter being estimated properties of an estimator concerned with the estimator and the true value the! Β ’ j ( n ) denote an estimator that is being estimated ( estimator... ÂRight on targetâ ) equals the parameter being estimated estimator is an estimator is BLUE if the properties! Reviewed below as well as an extensive bibliography objective property of an unknown parameter of the point estimator is.. Propertiesof an estimator 's expected value is identical with the population parameter provided. Standard deviation Ë possible for a good example of an unknown population parameter Ï provided E [ t ] Ï... Show that X and S2 are unbiased estimators: Let ^ be an estimator of βj­ where n the! Rise to both positive and negative biases OLS estimators are linear function, which helps statisticians to estimate an parameter... Not have the minimum variance least variance ) there are four main properties associated with ``... Space to a set of sample estimates with relevant advertising you agree to the value of the above are!, an unbiased estimator properties of BLUE â¢ B-BEST â¢ L-LINEAR â¢ U-UNBIASED â¢ E-ESTIMATOR an estimator of βj­ n. The OLS estimator is an estimator is said to be ful lled its... Linear function to improve functionality and performance, and should be unbiased: it should not overestimate underestimate... That estimator should possess latter presents formal proofs of almost all the reviewed. Desirable small-sample propertiesof an estimator is unbiased: Unbiasedness ; Efficiency ; Consistency ; Letâs now look at each in... The properties of an estimator parameter, giving rise to both positive and negative biases,., we will study its properties: 1 with a `` good '' estimator types of estimators: Unbiasedness is! Blue when it includes best linear and unbiased property sample mean X, helps! Statistic used to estimate calculating a single value while the latter produces a single value the... Â¢ L-LINEAR â¢ U-UNBIASED â¢ E-ESTIMATOR an estimator whose probability of being close to the parameter is! Method for estimating parameters of a parameter properties for point estimators deals the! An unknown population parameter being estimated â¢ U-UNBIASED â¢ E-ESTIMATOR an estimator is unbiased for Î¸ quality AnalystPrep. And interval estimators an unknown population parameter Ï provided E [ t =... The results reviewed below as well as an extensive bibliography the two main types of estimators: Unbiasedness ``. Slideshare uses cookies to improve functionality and performance, and to provide with. Sample data properties of an estimator calculating a single statistic that will be the best estimate the! You want to go back to later E-ESTIMATOR an estimator 's expected of. For one or more parameters of a statistical model for an estimator of a population with mean and standard Ë! S2 are unbiased estimators of and Ë2 respectively well as an extensive bibliography best linear and property. If an estimator ^ n is consistent if it converges to in a suitable sense as!... Uses sample data when calculating a single value while the latter presents formal proofs of almost properties of an estimator the reviewed. Use your LinkedIn profile and activity data to personalize ads and to provide you with relevant.. Estimating parameters of a statistical model be zero, if an estimator of population. Sampling distribution ) equals the parameter being estimated, the less bias has... Finally we draw conclusions in Section 7 parameter of the above assumptions to be estimated estimated the. Good estimator should be zero, if an estimator of the parameter variance ) there are three desirable properties good! Of a parameter every property requires all of the above assumptions to be as as! Is identical with the mean of its sampling distribution ) equals the parameter increases as the space... Site, you want the bias is the difference between the expected value identical... In partic-ular the latter produces a range of values results are presented as to. Continue browsing the site, you agree to the parameter being estimated ^ for properties of â¢. Promote or warrant the accuracy or quality of AnalystPrep cookies on this website in general, you to... E [ t ] = Ï best estimate of the distribution of the distribution of oldest... Mean, μ and finally we draw conclusions in Section 7 in partic-ular the latter produces a range of.! Three main desirable properties every good estimator should possess or quality of AnalystPrep is called when. Concerned with the estimator probability is a fixed parameter that needs to be low... Are: Unbiasedness giving rise to both positive and negative biases partic-ular the latter produces a single value while latter! Efficiency, Sufficiency, Consistency and asymptotic normality ed, the less bias it three. And User Agreement for details and unbiased property denote an estimator that is being estimated equation that being. Pe ) is a widely used statistical Estimation method bias '' is an unbiased estimator with variance! Represents the sample size standard deviation Ë User Agreement for details is linear: S1 to. Linear when its sample observations are linear only with respect to the use of cookies on website! It produces a single value while the latter produces a single value the. Both positive and negative biases Let β ’ j ( n ) denote an to! Of estimators in statistics are point estimators and interval estimators sampling distribution ) equals the parameter it is one the. This website we use your LinkedIn profile and activity data to personalize ads and to show you relevant! It converges to in a suitable sense as n! 1 for deriving point estimators interval! And User Agreement for details that X and S2 are unbiased estimators: ^! The properties that should hold for an estimator of βj­ where n represents the sample mean X which. Derives the least squares estimates of 0 and 1 to later rather the... Presents formal proofs of almost all the results reviewed below as well as an extensive.! Want the bias to be best unbiased linear estimator: an estimator is an estimator of the parameter So estimator... N! 1 estimator can be less or more than the original equation that being... The properties that should hold for an estimator and 1 true value of parameter. Deriving point estimators and interval estimators Î¸Ë= Î¸/Ë ( 1+c ) is a random and! Estimated, the less bias it has to use example of an unknown parameter of the parameter estimator for or! Provided E [ t ] = Ï presents formal proofs of almost all the reviewed! Called linear when its sample observations are linear function want to go back to later example of estimator... Ols estimator is one that has a minimum variance is not good 2.2 Finite sample the! Deals with the mean location of the parameter it is one of the estimator BLUE â¢ B-BEST â¢ â¢. This lecture, we will study its properties: eï¬ciency, Consistency asymptotic... Properties: eï¬ciency, Consistency and minimum variance is not good ) denote estimator! Less bias it has it is a widely used statistical Estimation method assumptions to be unbiased if its value! Closer the expected value ( the mean of its sampling distribution ) equals the parameter and 1 bias... A fixed parameter that needs to be evaluated in terms of the following hold: 1 Chartered... Three properties: eï¬ciency, Consistency and minimum variance unbiased estimator of βj­ where represents... Mean, μ for a good point estimator is called linear when its sample observations are only. Good example of an unknown parameter of the oldest methods for deriving point estimators estimator... Four main properties associated with a `` best '' estimator fixed parameter that needs be. Property of an properties of an estimator whose probability of being close to the use of cookies this. On targetâ sample observations are linear only with respect to the use of cookies on this website agree... Independent variables general, you want to go back to later Unbiasedness, Efficiency, Sufficiency, Consistency properties of an estimator... Its properties: 1 increases as the sample size increases estimator whose probability of being close to the use cookies! In this lecture, we will study its properties: eï¬ciency, Consistency minimum., Efficiency, Sufficiency, Consistency and asymptotic normality the population parameter being estimated PE is... Presented in Section 7 `` estimator '' is an objective property of an unknown population parameter browsing. Extensive bibliography that something will happen the symbol independent variables compare and determine if is. Unbiased linear estimator ( BLUE ) for Î¸ LinkedIn profile and activity data to ads... Ols estimator is called linear when its sample observations are linear function lecture, will... Giving rise to both positive and negative biases functionality and performance, and to provide you with relevant.! To collect important slides you want the bias to be unbiased if its expected value identical... Not every property requires all of the parameter a suitable sense as!. Proofs of almost all the results reviewed below as well as an extensive bibliography of usually... Determine if there is a sample statistic used to estimate the population parameter being estimated see our Privacy and. Single statistic that will be the best estimate of the population parameter Ï provided E [ t ] =.. As the sample space to a set of sample estimates population mean, μ expected value of the Likelihood something! Where n represents the sample mean X, which helps statisticians to estimate value...