Estimating Population Proportions by Means of Calibration Estimators

Size: px
Start display at page:

Download "Estimating Population Proportions by Means of Calibration Estimators"

Transcription

1 Revista Colombiaa de Estadística Jauary 2015, Volume 38, Issue 1, pp. 267 to 293 DOI: Estimatig Populatio Proportios by Meas of Calibratio Estimators Estimació de proporcioes poblacioales mediate estimadores de calibració Sergio Martíez 1,a, Atoio Arcos 2,b, Helea Martíez 1,c, Sarjider Sigh 3,d 1 Math Departmet, Uiversity of Almería, Almería, España 2 Departmet of Statistics ad Operatioal Research, Uiversity of Graada, Graada, España 3 Departmet of Mathematics, Texas A&M Uiversity-Kigsville, Kigsville Texas, Uited States Abstract This paper cosiders the problem of estimatig the populatio proportio of a categorical variable usig the calibratio framework. Differet situatios are explored accordig to the level of auxiliary iformatio available ad the theoretical properties are ivestigated. A ew class of estimator based upo the proposed calibratio estimators is also defied, ad the optimal estimator i the class, i the sese of miimal variace, is derived. Fially, a estimator of the populatio proportio, uder ew calibratio coditios, is defied. Simulatio studies are cosidered to evaluate the performace of the proposed calibratio estimators via the empirical relative bias ad the empirical relative efficiecy, ad favourable results are achieved. Key words: Auxiliary Iformatio, Calibratio, Estimators, Fiite Populatio, Samplig Desig. Resume El artículo cosidera el problema de la estimació de la proporció poblacioal de ua variable categórica usado como marco de trabajo la calibració. Se explora diferetes situacioes de acuerdo co la iformació auxiliar dispoible y se ivestiga las propiedades teóricas. Ua ueva clase de estimadores basada e los estimadores de calibració propuestos tambié a Professor. spuertas@ual.es b Professor. arcos@ugr.es c Ph.D. Research Assistat. hmartiez@ual.es d Associate Professor. sarjider.sigh@tamuk.edu 267

2 268 Sergio Martíez, Atoio Arcos, Helea Martíez & Sarjider Sigh es defiida y el estimador óptimo e la clase, e el setido de variaza míima, es obteido. Fialmete, u estimador de la proporció poblacioal, bajo uevas codicioes de calibració es tambié propuesto. Estudios de simulació para evaluar el comportamieto de los estimadores calibrados propuestos a través del sesgo relativo empírico y de la eficiecia relativa empírica so icluidos, obteiédose resultados satisfactorios. Palabras clave: calibració, diseño muestral, estimadores, iformació auxiliar, població fiita. 1. Itroductio I the presece of auxiliary iformatio, various approaches may be used to improve the precisio of estimators at the estimatio stage. The book of Sigh 2003 cotais several examples, icludig ratio, differece or calibratio estimators, followig the methodology proposed by Deville & Särdal 1992 ad Särdal 2007, or regressio estimators, as the papers of Arab, Shagodoyi & Sigh 2010 ad Sigh, Sigh & Kozak 2008 show. These techiques are geerally more efficiet tha other methods ot usig auxiliary iformatio. Usually social surveys are focused o categorical variables as sex, race, potetial voters, etc. Efficiet isertio of available auxiliary iformatio would improve the precisio the estimatios for the proportio of a categorical variable of iterest. Coceptually, it is difficult to justify usig a regressio estimator for estimatig proportios. Duchese 2003 cosidered estimators of a proportio uder differet samplig schemes ad preseted a estimator which used the logistic regressio estimator. The model calibratio techique proposed by Wu & Sitter 2001 ca be also used to estimate a proportio by usig a logistic regressio model. Based o logistic models, these estimators efficietly facilitate good modelig of survey data assumig that uit-specific auxiliary data i the populatio U are available. I this case it is assumed that the values of auxiliary variables are kow for the etire fiite populatio referred to as complete auxiliary iformatio but the values of mai variable are kow oly if the uit is selected i the sample. It is very commo for populatio data associated with auxiliary variables to be obtaied from cesus results, admiistrative files, etc., ad these sources ofte provide differet parameters for these auxiliary variables. For example, positio measures mea, media ad other momets are ormally provided, but there is o access to data for each idividual. I the preset study, it is assumed that the oly datum kow is the proportio of idividuals presetig oe or more characteristics related to the study variable. Uder this assumptio, Rueda, Muñoz, Arcos, Álvarez & Martíez 2011 defied a estimator ad various cofidece itervals for a proportio usig the ratio method. The results of their simulatio studies show that ratio estimators are more efficiet tha traditioal estimators. Cofidece itervals outperform alterative methods, especially i terms of iterval width. Revista Colombiaa de Estadística

3 Estimatig Proportios by Calibratio Estimators 269 Calibratio techiques were first employed by Deville & Särdal 1992 to estimate the total populatio, but this approach is also applicable to the estimatio more complex of parameters tha the total populatio. Relevat papers estimatig populatio variaces are Sigh 2001, Sigh, Hor, Chowdhury & Yu 1999 ad Farrell & Sigh The estimatio of fiite populatio distributio fuctios is studied i papers by Harms & Duchese 2006 or Rueda, Martíez, Martíez & Arcos 2007 ad the estimatio of quatiles i Rueda, Martíez- Puertas, Martíez-Puertas & Arcos I Sectio 2 we review a proportio estimator usig the calibratio techique. Sectio 3 describes alterative methods for derivig the calibratio estimator for the proposed parameter. I Sectio 4 we exted these methods to the multiple case. A simulatio study is performed i Sectio 5 ad our coclusios are reported i Sectio Calibratio Estimators for the Proportio 2.1. Defiitio of the Calibratio Estimator Assume a sample s with size from a fiite populatio U = {1, 2,..., } with size, selected by a specific samplig desig d, with iclusio probabilities π k ad π kl assumed to be strictly positive. Let A be a attribute of study i the populatio U, defiig A k = 1 whe a uit k of the populatio U has the attribute A ad A k = 0 otherwise. The populatio proportio of attribute A i the populatio U is give by: P A = 1 A k. 1 To estimate 1, the usual desig-weighted Horvitz-Thompso estimator is: where d k = 1/π k. P AH = 1 d k A k 2 If we cosider a auxiliary attribute B i which the value B k is kow for every uit k i the sample s ad is also kow, the above estimator caot icorporate the iformatio provided by the attribute B, i estimatig the populatio proportio of A. Oe way of icorporatig auxiliary iformatio i the parameter estimatio is via replacig the weights d k by ew weights ω k, usig calibratio techiques. Calibratio is a highly desirable property for survey weights, as Särdal 2007 argues, for the followig reasos: it provides a systematic way of takig auxiliary iformatio ito accout; it is a meas of obtaiig cosistet estimates, with kow aggregates; Revista Colombiaa de Estadística

4 270 Sergio Martíez, Atoio Arcos, Helea Martíez & Sarjider Sigh it is used by statistical agecies for estimatig differet fiite populatio parameters. Several atioal statistical agecies have developed software desiged to compute calibratio weights based o auxiliary iformatio available i populatio registers ad other sources. Such agecies iclude CLA Statistics Swede ad BASCULA Cetral Bureau of Statistics, The etherlads. Followig Deville & Särdal 1992 to obtai a calibratio estimator for the attribute A based o the attribute B, we calculate the weights ω k miimizig the chi-square distace χ 2 = ω k d k 2 3 d k q k subject to the coditio = 1 B k = 1 ω k B k 4 where q k are kow positive costats urelated to d k ad 0 < < 1. By miimizig 3 uder 4, the ew weights ω k are give by: ω k = d k + λd kq k B k where λ is the followig Lagrage multiplier λ = 2 H d k q k B k 5 ad H is the usual Horvitz-Thompso estimator for the attribute B. With the calibratio weights 5, assumig d k q k B k 0, the resultig estimator is: P AW = 1 ω k A k = P AH + H d k q k B k A k 6 d k q k B k By 4, whe the estimator is applied to estimate the populatio proportio of B, it coicides with Properties of the Calibratio Estimator Followig Deville & Särdal 1992, it ca be show that the estimator P AW is a asymptotically ubiased estimator for P A ad its asymptotic variace is give by AV P AW = 1 2 kl d k E k d l E l 7 l U Revista Colombiaa de Estadística

5 Estimatig Proportios by Calibratio Estimators 271 where kl = π kl π k π l ; E k = A k D B k ad D = q k B k A k q k B k A estimator for this variace is V P AW = kl d k e k d l e l 8 π kl l U d k q k B k A k with e k = A k B k D ad D = d k q k B k Example 1. Uder SRSWOR ad q k = 1 for all k U the estimator P AW is: where p A = 1 A k ; ad the asymptotic variace is P AW = p A + p B p p B p B = 1 B k ad p = 1 A k B k AV P AW = V P AV W = V p A + D 2 V p B 2DCov p A, p B [ 2 1 f = P A Q A P P A where Q A = 1 P A ; = 1, P = 1 ca be estimated by [ V P AW = 1 f p A q A + 1 p p B 2 p B q B 2 with q A = 1 1 A k ad q B = 1 1 B k 9 A k B k ad f =. This variace p 3. Alterative Calibratio Estimators p B p p A p B The usual estimator uder SRSWOR, p A has the followig shift ivariace property p A = 1 q A Revista Colombiaa de Estadística

6 272 Sergio Martíez, Atoio Arcos, Helea Martíez & Sarjider Sigh Hece p A has the same performace i the estimatio of P A as the performace of q A i the estimatio of Q A. I geeral, this property is ot satisfied by P AW. It is easy to see that this property is fulfilled if 1 = 1 ω k 12 Thus, we have two ways of obtaiig a estimator with the above property: i By cosiderig a calibratio estimator Q AW for Q A based o, ad determiig whe the estimator P AW has a smaller variace tha the estimator Q AW i order to defie a ew estimator based o these two. ii By cosiderig a calibratio estimator for P A based o ad because if we derive a calibratio estimator that provides perfect estimates for ad, the: 1 = + = 1 ω k B k + 1 ω k 1 B k = 1 ω k 3.1. A Estimator Based o the Complemetary: The P AT Estimator to The first alterative is developed oly uder SRSWOR, miimizig 3 subject = 1 = 1 ω k 1 B k 13 The resultig estimator, assumig q B 0, ca be expressed by with Q AW = q A + q B q q B 14 q = 1 1 B k 1 A k I the same way as with the estimator P AW i Example 1, the asymptotic variace of Q AW is give by: [ 2 AV Q 1 f Q AW = P A Q A Q 2 Q Q A where Q = 1 1 B k 1 A k. Revista Colombiaa de Estadística

7 Estimatig Proportios by Calibratio Estimators 273 A estimator V Q AW for 15 ca be easily defied by [ V Q AW = 1 f p A q A + 1 q q B 2 p B q B 2 q q B q q A q B 16 Let us ow compare the asymptotic variace of P AW with the followig estimator of P A, P AQ = 1 q. We have see Appedix A AV P AW < AV P AQ whe P AT = P < Q. 17 Hece, asymptotically, a more efficiet estimator for the populatio proportio P A is P AW if p < q or q B = 0 p B q B P AQ if p p B ote that the asymptotic variace of P AT is AV P AW AV P AT = AV P AQ q q B or p B = 0 if P otherwise < Q 18 To estimate the asymptotic variace of P AT 10 whe the followig coditio is satisfied the estimator ca be defied by p p B < q q B or q B = 0 19 ad it ca be defied by 16 if p p B q q B or p B = Whe i coditio 20 the equality is satisfied, we have P AW = P AQ, which implies that AV P AW = AV P AQ ad V P AW = V P AQ. The reaso for defiig P AT is to obtai a estimator with the property P AT = 1 Q AT. Accordigly, the estimator Q AT is defied by Q AT = Q AW if q q B 1 P AW f q q B < p p B or p B = 0 p p B or q B = 0 Revista Colombiaa de Estadística

8 274 Sergio Martíez, Atoio Arcos, Helea Martíez & Sarjider Sigh ad 1 Q AT = 1 Q AW if q q B P AW if q q B < p p B or p B = 0 p p B or q B = 0 Sice P AW = 1 Q AW whe q q B = p p B, we have 1 Q AT = P AT. Aother importat questio relative to the estimator P AT is: for low proportios, estimators such as P AW caot be calculated whe p B = 0. The estimator P AT does ot preset this problem, because if p B = 0 the estimator P AT = 1 Q AW ad sice q B = 1 we have P AT = 1 Q AW = 1 q A + q B q The the estimator P AT ca be obtaied for low proportios. Fially, the estimator P AT has the followig drawback: by 9; 15 ad 18, its asymptotic behaviour is worse tha that of the usual estimator whe P P A = Q Q A < 0 21 Thus, if coditio 21 occurs, we propose to use the attribute B = B c because P A B P A P B = 1 A k 1 B k P A 1 = = P A P P A + P A = P + P A > 0 Therefore, with attribute B, the above problem, whe 21 occurs, is solved A Optimal Estimator: The P Aα Estimator Aother way to improve the asymptotic behaviour is as follows: let us defie P Aα = α P AW + 1 α1 Q AW 22 ad take the value α that miimizes the variace. The miimum variace of P Aα is give by see Appedix B AV P Aα = G P A Q A φ2 with G = 1 f 1 ad φ = P P A = Q Q A Revista Colombiaa de Estadística

9 Estimatig Proportios by Calibratio Estimators 275 ad this is achieved where P α = P P A Q Q = The estimator P Aα has the desirable property because if we defie the P Aα = 1 Q Aα Q Aα = β Q AW + 1 β1 P AW Q Q Aα = 1 β P AW + β 1 Q AW = α P AW + 1 α1 Q AW with α = 1 β. Sice AV 1 Q Aα = AV Q Aα, we fid that miimizig the variace of Q Aα with respect to β is equal to miimizig the variace 1 Q Aα with respect to α, ad cosequetly α is agai give by 44 ad P Aα = 1 Q Aα. The P Aα estimator presets the followig disadvatages: It caot be calculated if p B = 0 or q B = 0 The optimum value α give by 44 depeds o theoretical variaces ad covariaces, which are geerally ukow. With respect to the secod questio, the value α ca be easily estimated whe the sample is draw by p p B α = p p B q p B q B Therefore, we obtai the followig estimator P Aα = α P AW + 1 α1 Q AW The estimator P Aα also has the desired property A estimator that Calibrate i ad at the Same Time: The P AR Estimator Whe usig the populatio size ad the populatio proportio of B, the auxiliary iformatio is the same as i the case of a post-stratified estimator. The secod way ii to obtai the property 11, based o a estimator for P A calibratio with ad, ca be developed i ay samplig desig. To do so, we must miimize the distace 3 uder the coditios Revista Colombiaa de Estadística

10 276 Sergio Martíez, Atoio Arcos, Helea Martíez & Sarjider Sigh = 1 ω k B k = 1 ω k 1 B k The calibratio weights with this ew calibratio process are 24 ω k = d k + H d k q k B k + H d k q k B k d k q k 1 B k d kq k 1 B k 25 ad the resultig estimator is where P AR = P AH + H B 1 + H B 2 26 B 1 = Therefore, whe d k q k B k A k ad B2 = d k q k B k d k q k B k = 0 d k q k 1 B k A k d k q k 1 B k or d k q k 1 B k = 0 the estimator 26 caot be obtaied. This problem is solved as follows: sice the two coditios are mutually exclusive, if d k q k B k = 0 we ca calibrate the estimator usig oly the attribute B. O the other had, if d k q k 1 B k = 0 the estimator P AR ca be developed oly with the attribute B. Thus, the estimator P AR is well defied. To prevet this article from becomig excessively log, i the same way as with the estimator P AW, the asymptotic variace of 26 is give by AV P AR = 1 2 l U 27 kl d k U k d l U l 28 Revista Colombiaa de Estadística

11 Estimatig Proportios by Calibratio Estimators 277 with U k = A k B 1 B K B 2 1 B k ad q k B k A k B 1 = ; B 2 = q k B k q k 1 B k A k q k 1 B k 28 is determied usig the followig estimator V P AR = 1 kl 2 d k u k d l u l 29 π kl l s where u k = A k B 1 B k B 2 1 B k Example 2. Uder SRSWOR ad q k = 1 for all k U, the estimator P AR ca be expressed by with P AR = p A + p B p p B p A B = 1 A k 1 B k. + q B p A B q B 30 I the same way as before with the estimator P AW, the asymptotic variace of P AR is see Appedix C. [ AV P 1 f P A 2 P A 2 AR = P A Q A [ P A 2 = G P A Q A = G P A Q A φ2 31 To estimate 45 the followig estimator is defied [ V P 1 f AR = p A q A φ 2 1 p B q b 32 with φ = p p A p B. Thus, the estimator P AR has the same asymptotic variace as the estimator P Aα. The, by 45, uder SRSWOR the estimator P AR is always more efficiet tha the estimators p A ad P AT. ote that the proposed estimator is essetially a post-stratified estimator ad, i this sese, is ot ew. Here, we look at it from a differet poit of view. I a practical situatio, the estimator P AR is preferred the estimator P Aα, sice P AR does ot eed estimatio of ay ukow populatio quatity. The case of estimator P Aα requires estimatig value α, which is geerally ukow. Revista Colombiaa de Estadística

12 278 Sergio Martíez, Atoio Arcos, Helea Martíez & Sarjider Sigh 4. Extesio to Multivariate Auxiliary Iformatio The previous sectio cosidered oly a auxiliary attribute B; let us ow assume that the study attribute A is related to J auxiliary attributes B 1,... B J. To develop the usual way of icorporatig the iformatio provided by J attributes i the estimatio of P A with calibratio techiques, we cosider a ew weight ω k subject to the followig coditios ext, we deote where P j = 1 B jk = 1 ω k B jk j = 1,..., J. 33 P = P 1,..., P J ; P H = P 1H,..., P JH ad B k = B 1k,..., B Jk P jh = 1 d k B jk j = 1,..., J. By T we deote the followig matrix T = d k q k B k B k. With the miimizatio of 3 uder the P coditios give by 33, the ew weights obtaied are: ω k = d k + d k q k P P H T 1 B k. 34 The calibratio estimator based o 34 is give by: with Ĥ = T 1 d k q k B k A k. P AW M = P AH + P P H Ĥ 35 ote that the weights 34 ad the estimator P AW M caot be obtaied if the matrix T is sigular. Followig Rueda et al. 2007a, the asymptotic variace of the estimator P AW M is AV P AW M = 1 2 kl d k Z k d l Z l 36 with Z k = A k B k H where l U 1 H = q k B k B k q k B k A k. Revista Colombiaa de Estadística

13 Estimatig Proportios by Calibratio Estimators 279 The asymptotic variace 36 ca be estimated by with z k = A k B k Ĥ V P AW M = 1 2 l s kl π kl d k z k d l z l 37 Example 3. Uder SRSWOR ad q k = 1, whe oly the two auxiliary attributes B ad C are cosidered, the matrix T ca be expressed by pb p T = BC p BC p C. Cosequetly, T = 2 p B p A p BC 2 ad pc p BC pc p BC T 1 = p BC T p B p BC p B = p B p C p BC 2. ext, we have p d k q k B k A k = p AC ad Ĥ = T 1 d k q k B k A k = pc p p p BC p B p AC p p BC p B p C p BC 2. Thus, the estimator P AW M, uder SRSWOR with two auxiliary attributes, is [ p C p p AC p BC P AW M = p A + p B p B p C p BC 2 [ 38 p B p AC p p BC + P C p C p B p C p BC 2 ow, if we deote A 1 = P CP P AC C P C C 2 ad A 2 = P AC P C P C C 2 Revista Colombiaa de Estadística

14 280 Sergio Martíez, Atoio Arcos, Helea Martíez & Sarjider Sigh The asymptotic variace of the estimator P AW M is AV P 1 f [ AW M = P A Q A + A A 2 2 P C Q C 1 + 2A 1 A 2 C P C 2A 1 P P A 1 f [ 2A 2 P AC P A P C = P A Q A A A 2 2 P C + 2A 1 A 2 C A 1 + A 2 P C 2 2A 1 P P A 2A 2 P AC P A P C. To determie 39 we use the followig estimator: V P 1 f [ AW M = p A q A + Â1 2 p B + Â2 2 p C + 2Â1Â2 p BC 1 Â1 p B + Â2 p C 2 2Â1 p p A p B 2Â2 p AC p A p C Simulatio study A limited study was carried out to ivestigate the desig-based fiite sample performace of the proposed estimators i compariso with that of covetioal estimators Simulated data The estimators were evaluated usig 15 simulated populatios with a populatio size = These populatios were geerated as a radom sample of 1000 uits from a Beroulli distributio with parameter P A = {0.5, 0.75, 0.9}, ad the attributes of iterest were thus achieved with the aforemetioed populatio proportios. Auxiliary attributes were also geerated, usig the same distributio, but a give proportio of values were radomly chaged so that Cramer s V coefficiet betwee the attribute of iterest ad the auxiliary attribute took the values 0.5, 0.6, 0.7, 0.8 ad 0.9. For each simulatio, 1000 samples with sizes = 50, 75, 100 ad 125, were selected uder SRSWOR to compare the estimators: 1 the Horvitz-Thompso estimator P AH 2 the ratio estimator P Aratio see Rueda et al., 2011, 3 P AW estimator, 4 P AT estimator, Revista Colombiaa de Estadística

15 Estimatig Proportios by Calibratio Estimators P Aα estimator with α estimated, 6 P AR estimator, 7 the multivariate ratio estimator P AratioM see Rueda et al., 2011 Multiple ratio, ad 8 the multivariate calibratio estimator P AW M. i terms of relative bias RB ad relative efficiecy with respect to the ratio estimator, where RB = E[ p A P A P A, = MSE[ p A MSE[ P Aratio, p A is a give estimator ad E[ ad MSE[ deote, respectively, the empirical mea ad the mea square error. Values of less tha 1 idicate that p A is more efficiet tha P Aratio. The results derived from this simulatio study gave values of RB withi a reasoable rage. All the calibratio estimators produced absolute relative bias values of less tha 1% except i case P A =0.9 ad φ=0.9. Uivariate ratio estimator produced the highest bias values, especially for small sample sizes. Figures 1, 2 ad 3 show the values of for the various populatios. These figures show: The ratio estimator performs poorly whe there is little associatio betwee the variables. Whe φ = 0.5 this estimator is worse tha the Horvitz- Thompso estimator. Eve whe φ = 0.6 as is sometimes the case P A = 0.75 ad P A = 0.9 the ratio estimator has a large MSE. I populatios with a large φ this problem does ot arise. With large φ values, all the estimators that use auxiliary iformatio produce good results: for φ 0.7 all calibratio ad ratio estimators are better tha the Horvitz-Thompso estimator. It is also see that as φ icreases, all the estimators achieve greater precisio, which is particularly marked for very high proportios. Of all the calibratio estimators, the first oe proposed P AW has the lowest degree of efficiecy. Although it performs better tha the Horvitz-Thompso estimator o most occasios except whe P A =0.9, φ=0.5 ad 0.6 the others produce a smaller MSE. The P AT, PAα ad P AR estimators perform very well i all cases. For high proportios P A =0.75 ad 0.9 the efficiecy of the estimators is fairly similar; oly i the case of P A =0.5 ad small values of φ is there a oticeable differece betwee them, i terms of efficiecy. I these cases, the best results are achieved by the P AR estimators that calibrate i ad at the same time. Revista Colombiaa de Estadística

16 282 Sergio Martíez, Atoio Arcos, Helea Martíez & Sarjider Sigh 1.0 φ = 0.5 Horvitz-Thompso 1.1 φ = 0.6 Horvitz-Thompso φ = 0.7 Horvitz-Thompso φ = 0.8 Horvitz-Thompso φ = 0.9 Horvitz-Thompso Figure 1: Empirical relative efficiecy of the differet estimators for the simulated populatios whe P A = 0.5 ad φ = 0.5, 0.6, 0.7, 0.8, 0.9. The sample size does ot produce a clear effect o the behaviour of the estimators; i some cases, as the sample size icreases, the efficiecy of the estimators icreases, while i others, it decreases as whe φ = 0.9 ad P A =0.9. ad calibratio estimators usig two auxiliary variables always have a lower tha those usig a sigle auxiliary variable. For P A =0.5 ad 0.5 φ 0.7 the multiple calibratio estimator is slightly more efficiet tha the multiple ratio estimator. For P A =0.75 ad 0.9 both estimators have similar levels of efficiecy. estimatio is usually kow to work well whe the variables auxiliary ad of iterest are positively correlated. I this case it is applied to 0-1 variables, so that a positive associatio is expected for the method to work higher freque- Revista Colombiaa de Estadística

17 Estimatig Proportios by Calibratio Estimators φ = 0.5 Horvitz-Thompso 1.1 φ = 0.6 Horvitz-Thompso φ = 0.7 Horvitz-Thompso 2.0 φ = 0.8 Horvitz-Thompso φ = Horvitz-Thompso Figure 2: Empirical relative efficiecy of the differet estimators for the simulated populatios whe P A=0.75 ad φ = 0.5, 0.6, 0.7, 0.8, 0.9. cies for the A=1; B=1 A=0; B=0 cases istead of the 0;1 1;0 cases. From this study we ca coclude that the associatio betwee the variables is the most importat factor ifluecig the behaviour of ratio ad also of calibratio estimators. As expected, as φ icreases the M SE of the calibrated estimators decreases. Eve for moderate values of φ the calibratio estimators improve cosiderably, i terms of efficiecy, o the Horvitz-Thompso estimator. The behaviour of calibratio estimators is similar for small proportios, whereas whe the proportio approaches 1 there are larger differeces amog the proposed calibratio estimators. Hovewer, it is ot a easy task to quatify how much associatio is eeded for a good improvemet i terms of efficiecy, or whe too small that it becomes harmful to itroduce extra variables i the calibratio costraits. Revista Colombiaa de Estadística

18 284 Sergio Martíez, Atoio Arcos, Helea Martíez & Sarjider Sigh 0.9 φ = 0.5 Horvitz-Thompso 0.9 φ = 0.6 Horvitz-Thompso φ = Horvitz-Thompso 2.0 φ = 0.8 Horvitz-Thompso φ = Horvitz-Thompso Figure 3: Empirical relative efficiecy of the differet estimators for the simulated populatios whe P A=0.9 ad φ = 0.5, 0.6, 0.7, 0.8, Real Data I this sectio we apply some proposed estimators to data obtaied i a survey o perceptios of immigratio i a certai regio i Spai. A sample of size = 1919 was selected from a populatio with size = , usig stratified radom samplig. Amog topics of iterest i the survey was estimatig the percetage of citizes who believe that the authorities should make immigratio more difficult by imposig stricter coditios. The auxiliary variable available is the respodet s geder. This variable was observed i the sample ad the totals are kow for each provice stratum. Revista Colombiaa de Estadística

19 Estimatig Proportios by Calibratio Estimators 285 Three mai variables are icluded i this study, related to goodess of immigratio ad amout of immigratio. The mai variables are the aswers to the followig questios: i geeral, do you thik that for Adalusia, immigratio is...? c1-very bad, c2 Bad, c3 either good or bad, c4 Good, c5 Very good, ad i relatio to the umber of immigrats curretly livig i Adalusia, do you thik there are...? c1-too may, c2-a reasoable umber, c3-too few. I this simulatio study, we use the sample as populatio ad we draw stratified radom samples of size = 240 with proportioal allocatio eight stratum. Relative efficiecy with respect to the ratio estimator is computed, as i the previous case, for compared estimators over 1000 simulatio rus. We computed this relative efficiecy for each category of the mai variable 5 categories i the first case, ad 3 i the secod case ad the average over categories is also computed. At the same time, cofidece itervals based o a ormal distributio ad usig proposed estimated variaces are computed for each proportio. Table 1 shows the average legth of the 1000 simulatio rus for each category ad the average over categories. I a similar way, the empirical coverage of the cofidece estimatio is computed. Tables 1 ad 2 show that, from a efficiecy stadpoit P Aα is best. Lookig the average legth of cofidece itervals for the proportio i each category, ad the average over categories, the best estimator is P AR, but the optimal PAα has very similar results. However, the empirical coverage of cofidece itervals is closer to the omial level whe the optimal estimator is used. 6. Applicatio IESA, the Istitute for Advaced Social Studies coducted a survey betwee Jauary 14th ad February 13th, 2011 o the perceptio of culture i the Spaish regio of Adalusia Barometer of Culture of Adalusia - BACU. It is based o a sample draw from a ladlie phoe frame = 5,064,304. From this frame a stratified radom sample without replacemet of dimesio = 641 was selected, where strata were made up by eight geographical regios. Strata populatio sizes are h = , , , , , , , ad the correspodig strata sample sizes are h = 53, 99, 66, 62, 38, 49, 131, 143. Amog several topics of iterest i the survey, is the iterest to estimate perceptio of their culture i relatio to Europea citizes. A auxiliary variable available is geder which totals are kow for each strata. From Table 3 we observe that the P Aα estimator produces the best cofidece itervals. Revista Colombiaa de Estadística

20 286 Sergio Martíez, Atoio Arcos, Helea Martíez & Sarjider Sigh Table 1: Relative efficiecy with respect to the ratio estimator; legth ad empirical coverage of 95% cofidece level estimatio of proportios. Mai variable: goodess of immigratio. Stratified radom samplig. Estimator c1 c2 c3 c4 c5 avg Relative Efficiecy P AW P AT P Aα P AR Legth P AW P AT P Aα P AR Coverage P AW P AT P Aα P AR Table 2: Relative efficiecy with respect to the ratio estimator; legth ad empirical coverage of 95% cofidece level estimatio of proportios. Mai variable: amout of immigratio. Stratified radom samplig. Estimator c1 c2 c3 avg Relative Efficiecy P AW P AT P Aα P AR Legth P AW P AT P Aα P AR Coverage P AW P AT P Aα P AR Revista Colombiaa de Estadística

21 Estimatig Proportios by Calibratio Estimators 287 Table 3: Estimated proportio ˆp, lower boud lb, upper boud ub ad legth l of a 95% cofidece iterval uder stratified radom samplig. Do you thik that i Adalusia the cultural level, compared to the Europea Uio, is...? Estimator prop lb ub le Much lower P AW P AT P AR P Aα lower P AW P AT P AR P Aα Equal P AW P AT P AR P Aα Higher P AW P AT P AR P Aα Much higher P AW P AT P AR P Aα Coclusios I practice, it is importat to make the best possible use of available auxiliary iformatio so as to obtai the most efficiet estimator possible. Whe a proportio ca be estimated i the case of complete auxiliary iformatio i.e., whe auxiliary iformatio is available at the populatio level for each uit it is possible to cosider estimators that use the logistic regressio model Duchese, 2003, Wu ad Sitter, 2001, as a improvemet o the simple estimator. Whe there is merely a rearragemet of the populatio proportio of a attribute with respect to the study variable, the traditioal idirect methods Revista Colombiaa de Estadística

22 288 Sergio Martíez, Atoio Arcos, Helea Martíez & Sarjider Sigh such as the ratio by Rueda et al or the calibratio studied i this work ca be applied. We have studied four calibratio estimators for the proportio which are simple to calculate from stadard calibratio packages ad ca give rise to cosiderable icreases i the precisio achieved, as illustrated by the theoretical results reported here ad by the simulatio performed. The proposed estimators P AW ad P AR ca be obtaied from ay arbitrary samplig desig, whereas P Aα ad P AT estimators are defied uder SRSWOR. However, the extesio to a stratified radom samplig is straightforward. Cofidece itervals based o the estimated variaces of the studied calibratio estimators is also ivestigated through a limited simulatio study, uder a more realistic survey stratified radom samplig usig real data. PAα ad P AR have good properties i cofidece estimatio, ad i some sese a balace betwee legth ad coverage the optimal estimator P Aα provides the best results. Ackowledgemets The authors thak the Editor ad the two aoymous referees of this joural for their helpful commets o a earlier versio of this paper. This study was partially supported by Miisterio de Educació y Ciecia grat MTM , Spai ad by Cosejería de Ecoomía, Iovació, Ciecia y Empleo grat SEJ2954, Juta de Adalucía. [ Received: October 2013 Accepted: ovember 2014 Refereces Arab, R., Shagodoyi, D. K. & Sigh, S. 2010, Variace estimatio of a geeralized regressio predictor, Joural of the Idia Society of Agricultural Statistics 642, Deville, J.-C. & Särdal, C.-E. 1992, Calibratio estimators i survey samplig, Joural of the America Statistical Associatio 87418, Duchese, P. 2003, Estimatio of a proportio with survey data, Joural of Statistics Educatio 113, Farrell, P. & Sigh, S. 2005, Model-assisted higher-order calibratio of estimators of variace, Australia & ew Zealad Joural of Statistics 473, Harms, T. & Duchese, P. 2006, O calibratio estimatio for quatiles, Survey Methodology 32, Rueda, M., Martíez-Puertas, S., Martíez-Puertas, H. & Arcos, A. 2007, Calibratio methods for estimatig quatiles, Metrika 663, Revista Colombiaa de Estadística

23 Estimatig Proportios by Calibratio Estimators 289 Rueda, M., Martíez, S., Martíez, H. & Arcos, A. 2007, Estimatio of the distributio fuctio with calibratio methods, Joural of Statistical Plaig ad Iferece 1372, Rueda, M., Muñoz, J. F., Arcos, A., Álvarez, E. & Martíez, S. 2011, Estimators ad cofidece itervals for the proportio usig biary auxiliary iformatio with applicatios to pharmaceutical studies, Joural of Biopharmaceutical Statistics 213, Särdal, C. 2007, The calibratio approach i survey theory ad practice, Survey Methodology 332, Sigh, H. P., Sigh, S. & Kozak, M. 2008, A family of estimators of fiitepopulatio distributio fuctios usig auxiliary iformatio, Acta Applicadae Mathematicae 1042, Sigh, S. 2001, Geeralized calibratio approach for estimatig variace i survey samplig, Aals of the Istitute of Statistical Mathematics 532, Sigh, S. 2003, Advaced Samplig Theory with Applicatios: How Michael selected Amy, Kluwer Academic Publishers. Sigh, S., Hor, S., Chowdhury, S. & Yu, F. 1999, Calibratio of the estimator of variace, Australia ad ew Zealad Joural of Statistics 41, Wu, C. & Sitter, Y. R. 2001, A model-calibratio approach to usig complete auxiliary iformatio from survey data, Jourals - America Statistical Associatio 96, Appedix A. Compariso betwee AV P AW ad AV P AQ Because AV P AQ = AV q we have AV P AW < AV P AQ whe [ P A Q A + or equivaletly P < [ P A Q A + Q P P A 2 2 P P A < Q2 2 Q Q Q Q A Q Q A. Revista Colombiaa de Estadística

24 290 Sergio Martíez, Atoio Arcos, Helea Martíez & Sarjider Sigh ow, we have Q Q A = 1 1 A k 1 B k 1 P A 1 = P P A. The AV P AW < AV P AQ whe 2 Q 2 2 Q P P A < 0 therefore Sice where Q [ + Q 2P P A = K 1 K 2 < 0. K 2 = P + P A = P ĀB + P A B 0 P ĀB = 1 1 A k B k ad P A B = 1 A k 1 B k, we deduce that AV P AW < AV P AQ whe k 1 < 0, that is P < Q. 41 Appedix B. Obtaiig the miimum variace of P Aα If we deote V 1 = AV P AW ; V 2 = AV Q AW ad C = Cov P AW, Q AW the miimum variace of P Aα is V 2 V 1 C 2 42 V 1 + V 2 + 2C ad this is achieved whe It is easy to see that α = V 2 + C V 1 + V 2 + 2C [ C = Cov P AW, Q 1 f AW = P A Q A Q P P A Q 43 Revista Colombiaa de Estadística

25 Estimatig Proportios by Calibratio Estimators 291 Therefore, we have: V 2 + C = 1 f [ P P A Q 1 Q 2 Q + = 1 f Q 1 [ P P A Q Similarly V 1 + V 2 + 2C = 1 f 1 [ 2 + = 1 f 1 Q 2 2 Q Q 2 By substitutig the values V 2 + C ad V 1 + V 2 2C i 43, the value of α is foud to be: P α = P P A Q Q = Q. 44 V 1 V 2 = G 2 [ PA Q A 2 + [ + Q Q 2 PB + 4 Q Q 2φ + Q P A Q A + P Q [ = G 2 P A Q A φ + Q 2 φ [ Q 2 + PB + Q + 2 Q P A Q A 2φ φ 2 Q 2 + Q P A Q A 2 P Q Revista Colombiaa de Estadística

26 292 Sergio Martíez, Atoio Arcos, Helea Martíez & Sarjider Sigh [ = G 2 P A Q A φ + Q 2 + P A Q A φ2 Q + 2P Q P A Q A φ + where K = P A Q A φ Q 2 PB = G 2 P A Q A + Q 2 2 Q 2. PB O the other had φ2 C 2 = G [φ 2 + Q P A Q A + Q 2 Q 2 + G 2 K 2 + 2P Q P A Q A φ + Q Q 2 = G 2 K Thus, by substitutig the values C 2 ; V 1 V 2 ad V 1 +V 2 +2C i 42, we have: AV P Aα = G P A Q A φ2 Appedix C. Obtaiig AV P AR [ AV P 1 f AR = P A Q A P P A 2 PA B P 2 A B P A B P A with P A B = 1 A k 1 B k. Revista Colombiaa de Estadística

27 Estimatig Proportios by Calibratio Estimators 293 ow, takig ito accout that P A B P A = P A P P A + P A = P A P the asymptotic variace is Sice AV P AR = P A B 1 f 1 [ P A Q A + PA B + 2P P A P. P = P A B + P P P 2 A B = P A P we have [ AV P 1 f P A 2 P A 2 AR = P AQ A [ P A 2 45 = G P AQ A = G P AQ A φ2 Revista Colombiaa de Estadística

7-1. Chapter 4. Part I. Sampling Distributions and Confidence Intervals

7-1. Chapter 4. Part I. Sampling Distributions and Confidence Intervals 7-1 Chapter 4 Part I. Samplig Distributios ad Cofidece Itervals 1 7- Sectio 1. Samplig Distributio 7-3 Usig Statistics Statistical Iferece: Predict ad forecast values of populatio parameters... Test hypotheses

More information

Random Variables, Sampling and Estimation

Random Variables, Sampling and Estimation Chapter 1 Radom Variables, Samplig ad Estimatio 1.1 Itroductio This chapter will cover the most importat basic statistical theory you eed i order to uderstad the ecoometric material that will be comig

More information

Element sampling: Part 2

Element sampling: Part 2 Chapter 4 Elemet samplig: Part 2 4.1 Itroductio We ow cosider uequal probability samplig desigs which is very popular i practice. I the uequal probability samplig, we ca improve the efficiecy of the resultig

More information

Statistical Inference (Chapter 10) Statistical inference = learn about a population based on the information provided by a sample.

Statistical Inference (Chapter 10) Statistical inference = learn about a population based on the information provided by a sample. Statistical Iferece (Chapter 10) Statistical iferece = lear about a populatio based o the iformatio provided by a sample. Populatio: The set of all values of a radom variable X of iterest. Characterized

More information

Econ 325 Notes on Point Estimator and Confidence Interval 1 By Hiro Kasahara

Econ 325 Notes on Point Estimator and Confidence Interval 1 By Hiro Kasahara Poit Estimator Eco 325 Notes o Poit Estimator ad Cofidece Iterval 1 By Hiro Kasahara Parameter, Estimator, ad Estimate The ormal probability desity fuctio is fully characterized by two costats: populatio

More information

Improved Class of Ratio -Cum- Product Estimators of Finite Population Mean in two Phase Sampling

Improved Class of Ratio -Cum- Product Estimators of Finite Population Mean in two Phase Sampling Global Joural of Sciece Frotier Research: F Mathematics ad Decisio Scieces Volume 4 Issue 2 Versio.0 Year 204 Type : Double Blid Peer Reviewed Iteratioal Research Joural Publisher: Global Jourals Ic. (USA

More information

Chain ratio-to-regression estimators in two-phase sampling in the presence of non-response

Chain ratio-to-regression estimators in two-phase sampling in the presence of non-response ProbStat Forum, Volume 08, July 015, Pages 95 10 ISS 0974-335 ProbStat Forum is a e-joural. For details please visit www.probstat.org.i Chai ratio-to-regressio estimators i two-phase samplig i the presece

More information

A General Family of Estimators for Estimating Population Variance Using Known Value of Some Population Parameter(s)

A General Family of Estimators for Estimating Population Variance Using Known Value of Some Population Parameter(s) Rajesh Sigh, Pakaj Chauha, Nirmala Sawa School of Statistics, DAVV, Idore (M.P.), Idia Floreti Smaradache Uiversity of New Meico, USA A Geeral Family of Estimators for Estimatig Populatio Variace Usig

More information

FACULTY OF MATHEMATICAL STUDIES MATHEMATICS FOR PART I ENGINEERING. Lectures

FACULTY OF MATHEMATICAL STUDIES MATHEMATICS FOR PART I ENGINEERING. Lectures FACULTY OF MATHEMATICAL STUDIES MATHEMATICS FOR PART I ENGINEERING Lectures MODULE 5 STATISTICS II. Mea ad stadard error of sample data. Biomial distributio. Normal distributio 4. Samplig 5. Cofidece itervals

More information

A statistical method to determine sample size to estimate characteristic value of soil parameters

A statistical method to determine sample size to estimate characteristic value of soil parameters A statistical method to determie sample size to estimate characteristic value of soil parameters Y. Hojo, B. Setiawa 2 ad M. Suzuki 3 Abstract Sample size is a importat factor to be cosidered i determiig

More information

Resampling Methods. X (1/2), i.e., Pr (X i m) = 1/2. We order the data: X (1) X (2) X (n). Define the sample median: ( n.

Resampling Methods. X (1/2), i.e., Pr (X i m) = 1/2. We order the data: X (1) X (2) X (n). Define the sample median: ( n. Jauary 1, 2019 Resamplig Methods Motivatio We have so may estimators with the property θ θ d N 0, σ 2 We ca also write θ a N θ, σ 2 /, where a meas approximately distributed as Oce we have a cosistet estimator

More information

Statistics 511 Additional Materials

Statistics 511 Additional Materials Cofidece Itervals o mu Statistics 511 Additioal Materials This topic officially moves us from probability to statistics. We begi to discuss makig ifereces about the populatio. Oe way to differetiate probability

More information

It should be unbiased, or approximately unbiased. Variance of the variance estimator should be small. That is, the variance estimator is stable.

It should be unbiased, or approximately unbiased. Variance of the variance estimator should be small. That is, the variance estimator is stable. Chapter 10 Variace Estimatio 10.1 Itroductio Variace estimatio is a importat practical problem i survey samplig. Variace estimates are used i two purposes. Oe is the aalytic purpose such as costructig

More information

Big Picture. 5. Data, Estimates, and Models: quantifying the accuracy of estimates.

Big Picture. 5. Data, Estimates, and Models: quantifying the accuracy of estimates. 5. Data, Estimates, ad Models: quatifyig the accuracy of estimates. 5. Estimatig a Normal Mea 5.2 The Distributio of the Normal Sample Mea 5.3 Normal data, cofidece iterval for, kow 5.4 Normal data, cofidece

More information

Double Stage Shrinkage Estimator of Two Parameters. Generalized Exponential Distribution

Double Stage Shrinkage Estimator of Two Parameters. Generalized Exponential Distribution Iteratioal Mathematical Forum, Vol., 3, o. 3, 3-53 HIKARI Ltd, www.m-hikari.com http://dx.doi.org/.9/imf.3.335 Double Stage Shrikage Estimator of Two Parameters Geeralized Expoetial Distributio Alaa M.

More information

Estimation of a population proportion March 23,

Estimation of a population proportion March 23, 1 Social Studies 201 Notes for March 23, 2005 Estimatio of a populatio proportio Sectio 8.5, p. 521. For the most part, we have dealt with meas ad stadard deviatios this semester. This sectio of the otes

More information

Estimation of the Population Mean in Presence of Non-Response

Estimation of the Population Mean in Presence of Non-Response Commuicatios of the Korea Statistical Society 0, Vol. 8, No. 4, 537 548 DOI: 0.535/CKSS.0.8.4.537 Estimatio of the Populatio Mea i Presece of No-Respose Suil Kumar,a, Sadeep Bhougal b a Departmet of Statistics,

More information

Confidence intervals summary Conservative and approximate confidence intervals for a binomial p Examples. MATH1005 Statistics. Lecture 24. M.

Confidence intervals summary Conservative and approximate confidence intervals for a binomial p Examples. MATH1005 Statistics. Lecture 24. M. MATH1005 Statistics Lecture 24 M. Stewart School of Mathematics ad Statistics Uiversity of Sydey Outlie Cofidece itervals summary Coservative ad approximate cofidece itervals for a biomial p The aïve iterval

More information

1 Inferential Methods for Correlation and Regression Analysis

1 Inferential Methods for Correlation and Regression Analysis 1 Iferetial Methods for Correlatio ad Regressio Aalysis I the chapter o Correlatio ad Regressio Aalysis tools for describig bivariate cotiuous data were itroduced. The sample Pearso Correlatio Coefficiet

More information

Topic 9: Sampling Distributions of Estimators

Topic 9: Sampling Distributions of Estimators Topic 9: Samplig Distributios of Estimators Course 003, 2016 Page 0 Samplig distributios of estimators Sice our estimators are statistics (particular fuctios of radom variables), their distributio ca be

More information

Estimation of Gumbel Parameters under Ranked Set Sampling

Estimation of Gumbel Parameters under Ranked Set Sampling Joural of Moder Applied Statistical Methods Volume 13 Issue 2 Article 11-2014 Estimatio of Gumbel Parameters uder Raked Set Samplig Omar M. Yousef Al Balqa' Applied Uiversity, Zarqa, Jorda, abuyaza_o@yahoo.com

More information

Frequentist Inference

Frequentist Inference Frequetist Iferece The topics of the ext three sectios are useful applicatios of the Cetral Limit Theorem. Without kowig aythig about the uderlyig distributio of a sequece of radom variables {X i }, for

More information

Econ 325/327 Notes on Sample Mean, Sample Proportion, Central Limit Theorem, Chi-square Distribution, Student s t distribution 1.

Econ 325/327 Notes on Sample Mean, Sample Proportion, Central Limit Theorem, Chi-square Distribution, Student s t distribution 1. Eco 325/327 Notes o Sample Mea, Sample Proportio, Cetral Limit Theorem, Chi-square Distributio, Studet s t distributio 1 Sample Mea By Hiro Kasahara We cosider a radom sample from a populatio. Defiitio

More information

Lecture 2: Monte Carlo Simulation

Lecture 2: Monte Carlo Simulation STAT/Q SCI 43: Itroductio to Resamplig ethods Sprig 27 Istructor: Ye-Chi Che Lecture 2: ote Carlo Simulatio 2 ote Carlo Itegratio Assume we wat to evaluate the followig itegratio: e x3 dx What ca we do?

More information

Properties and Hypothesis Testing

Properties and Hypothesis Testing Chapter 3 Properties ad Hypothesis Testig 3.1 Types of data The regressio techiques developed i previous chapters ca be applied to three differet kids of data. 1. Cross-sectioal data. 2. Time series data.

More information

32 estimating the cumulative distribution function

32 estimating the cumulative distribution function 32 estimatig the cumulative distributio fuctio 4.6 types of cofidece itervals/bads Let F be a class of distributio fuctios F ad let θ be some quatity of iterest, such as the mea of F or the whole fuctio

More information

Discrete Mathematics for CS Spring 2008 David Wagner Note 22

Discrete Mathematics for CS Spring 2008 David Wagner Note 22 CS 70 Discrete Mathematics for CS Sprig 2008 David Wager Note 22 I.I.D. Radom Variables Estimatig the bias of a coi Questio: We wat to estimate the proportio p of Democrats i the US populatio, by takig

More information

Simulation. Two Rule For Inverting A Distribution Function

Simulation. Two Rule For Inverting A Distribution Function Simulatio Two Rule For Ivertig A Distributio Fuctio Rule 1. If F(x) = u is costat o a iterval [x 1, x 2 ), the the uiform value u is mapped oto x 2 through the iversio process. Rule 2. If there is a jump

More information

This is an introductory course in Analysis of Variance and Design of Experiments.

This is an introductory course in Analysis of Variance and Design of Experiments. 1 Notes for M 384E, Wedesday, Jauary 21, 2009 (Please ote: I will ot pass out hard-copy class otes i future classes. If there are writte class otes, they will be posted o the web by the ight before class

More information

MATH 320: Probability and Statistics 9. Estimation and Testing of Parameters. Readings: Pruim, Chapter 4

MATH 320: Probability and Statistics 9. Estimation and Testing of Parameters. Readings: Pruim, Chapter 4 MATH 30: Probability ad Statistics 9. Estimatio ad Testig of Parameters Estimatio ad Testig of Parameters We have bee dealig situatios i which we have full kowledge of the distributio of a radom variable.

More information

Tests of Hypotheses Based on a Single Sample (Devore Chapter Eight)

Tests of Hypotheses Based on a Single Sample (Devore Chapter Eight) Tests of Hypotheses Based o a Sigle Sample Devore Chapter Eight MATH-252-01: Probability ad Statistics II Sprig 2018 Cotets 1 Hypothesis Tests illustrated with z-tests 1 1.1 Overview of Hypothesis Testig..........

More information

Topic 9: Sampling Distributions of Estimators

Topic 9: Sampling Distributions of Estimators Topic 9: Samplig Distributios of Estimators Course 003, 2018 Page 0 Samplig distributios of estimators Sice our estimators are statistics (particular fuctios of radom variables), their distributio ca be

More information

Efficient GMM LECTURE 12 GMM II

Efficient GMM LECTURE 12 GMM II DECEMBER 1 010 LECTURE 1 II Efficiet The estimator depeds o the choice of the weight matrix A. The efficiet estimator is the oe that has the smallest asymptotic variace amog all estimators defied by differet

More information

Computing Confidence Intervals for Sample Data

Computing Confidence Intervals for Sample Data Computig Cofidece Itervals for Sample Data Topics Use of Statistics Sources of errors Accuracy, precisio, resolutio A mathematical model of errors Cofidece itervals For meas For variaces For proportios

More information

Topic 9: Sampling Distributions of Estimators

Topic 9: Sampling Distributions of Estimators Topic 9: Samplig Distributios of Estimators Course 003, 2018 Page 0 Samplig distributios of estimators Sice our estimators are statistics (particular fuctios of radom variables), their distributio ca be

More information

Analysis of Experimental Data

Analysis of Experimental Data Aalysis of Experimetal Data 6544597.0479 ± 0.000005 g Quatitative Ucertaity Accuracy vs. Precisio Whe we make a measuremet i the laboratory, we eed to kow how good it is. We wat our measuremets to be both

More information

On stratified randomized response sampling

On stratified randomized response sampling Model Assisted Statistics ad Applicatios 1 (005,006) 31 36 31 IOS ress O stratified radomized respose samplig Jea-Bok Ryu a,, Jog-Mi Kim b, Tae-Youg Heo c ad Chu Gu ark d a Statistics, Divisio of Life

More information

G. R. Pasha Department of Statistics Bahauddin Zakariya University Multan, Pakistan

G. R. Pasha Department of Statistics Bahauddin Zakariya University Multan, Pakistan Deviatio of the Variaces of Classical Estimators ad Negative Iteger Momet Estimator from Miimum Variace Boud with Referece to Maxwell Distributio G. R. Pasha Departmet of Statistics Bahauddi Zakariya Uiversity

More information

Simple Random Sampling!

Simple Random Sampling! Simple Radom Samplig! Professor Ro Fricker! Naval Postgraduate School! Moterey, Califoria! Readig:! 3/26/13 Scheaffer et al. chapter 4! 1 Goals for this Lecture! Defie simple radom samplig (SRS) ad discuss

More information

Approximate Confidence Interval for the Reciprocal of a Normal Mean with a Known Coefficient of Variation

Approximate Confidence Interval for the Reciprocal of a Normal Mean with a Known Coefficient of Variation Metodološki zvezki, Vol. 13, No., 016, 117-130 Approximate Cofidece Iterval for the Reciprocal of a Normal Mea with a Kow Coefficiet of Variatio Wararit Paichkitkosolkul 1 Abstract A approximate cofidece

More information

Stat 421-SP2012 Interval Estimation Section

Stat 421-SP2012 Interval Estimation Section Stat 41-SP01 Iterval Estimatio Sectio 11.1-11. We ow uderstad (Chapter 10) how to fid poit estimators of a ukow parameter. o However, a poit estimate does ot provide ay iformatio about the ucertaity (possible

More information

A quick activity - Central Limit Theorem and Proportions. Lecture 21: Testing Proportions. Results from the GSS. Statistics and the General Population

A quick activity - Central Limit Theorem and Proportions. Lecture 21: Testing Proportions. Results from the GSS. Statistics and the General Population A quick activity - Cetral Limit Theorem ad Proportios Lecture 21: Testig Proportios Statistics 10 Coli Rudel Flip a coi 30 times this is goig to get loud! Record the umber of heads you obtaied ad calculate

More information

Statistical inference: example 1. Inferential Statistics

Statistical inference: example 1. Inferential Statistics Statistical iferece: example 1 Iferetial Statistics POPULATION SAMPLE A clothig store chai regularly buys from a supplier large quatities of a certai piece of clothig. Each item ca be classified either

More information

Output Analysis (2, Chapters 10 &11 Law)

Output Analysis (2, Chapters 10 &11 Law) B. Maddah ENMG 6 Simulatio Output Aalysis (, Chapters 10 &11 Law) Comparig alterative system cofiguratio Sice the output of a simulatio is radom, the comparig differet systems via simulatio should be doe

More information

Topic 10: Introduction to Estimation

Topic 10: Introduction to Estimation Topic 0: Itroductio to Estimatio Jue, 0 Itroductio I the simplest possible terms, the goal of estimatio theory is to aswer the questio: What is that umber? What is the legth, the reactio rate, the fractio

More information

6 Sample Size Calculations

6 Sample Size Calculations 6 Sample Size Calculatios Oe of the major resposibilities of a cliical trial statisticia is to aid the ivestigators i determiig the sample size required to coduct a study The most commo procedure for determiig

More information

MBACATÓLICA. Quantitative Methods. Faculdade de Ciências Económicas e Empresariais UNIVERSIDADE CATÓLICA PORTUGUESA 9. SAMPLING DISTRIBUTIONS

MBACATÓLICA. Quantitative Methods. Faculdade de Ciências Económicas e Empresariais UNIVERSIDADE CATÓLICA PORTUGUESA 9. SAMPLING DISTRIBUTIONS MBACATÓLICA Quatitative Methods Miguel Gouveia Mauel Leite Moteiro Faculdade de Ciêcias Ecoómicas e Empresariais UNIVERSIDADE CATÓLICA PORTUGUESA 9. SAMPLING DISTRIBUTIONS MBACatólica 006/07 Métodos Quatitativos

More information

Lecture 3. Properties of Summary Statistics: Sampling Distribution

Lecture 3. Properties of Summary Statistics: Sampling Distribution Lecture 3 Properties of Summary Statistics: Samplig Distributio Mai Theme How ca we use math to justify that our umerical summaries from the sample are good summaries of the populatio? Lecture Summary

More information

ECE 901 Lecture 12: Complexity Regularization and the Squared Loss

ECE 901 Lecture 12: Complexity Regularization and the Squared Loss ECE 90 Lecture : Complexity Regularizatio ad the Squared Loss R. Nowak 5/7/009 I the previous lectures we made use of the Cheroff/Hoeffdig bouds for our aalysis of classifier errors. Hoeffdig s iequality

More information

ACCESS TO SCIENCE, ENGINEERING AND AGRICULTURE: MATHEMATICS 1 MATH00030 SEMESTER / Statistics

ACCESS TO SCIENCE, ENGINEERING AND AGRICULTURE: MATHEMATICS 1 MATH00030 SEMESTER / Statistics ACCESS TO SCIENCE, ENGINEERING AND AGRICULTURE: MATHEMATICS 1 MATH00030 SEMESTER 1 018/019 DR. ANTHONY BROWN 8. Statistics 8.1. Measures of Cetre: Mea, Media ad Mode. If we have a series of umbers the

More information

Chapter 8: STATISTICAL INTERVALS FOR A SINGLE SAMPLE. Part 3: Summary of CI for µ Confidence Interval for a Population Proportion p

Chapter 8: STATISTICAL INTERVALS FOR A SINGLE SAMPLE. Part 3: Summary of CI for µ Confidence Interval for a Population Proportion p Chapter 8: STATISTICAL INTERVALS FOR A SINGLE SAMPLE Part 3: Summary of CI for µ Cofidece Iterval for a Populatio Proportio p Sectio 8-4 Summary for creatig a 100(1-α)% CI for µ: Whe σ 2 is kow ad paret

More information

CHAPTER 8 FUNDAMENTAL SAMPLING DISTRIBUTIONS AND DATA DESCRIPTIONS. 8.1 Random Sampling. 8.2 Some Important Statistics

CHAPTER 8 FUNDAMENTAL SAMPLING DISTRIBUTIONS AND DATA DESCRIPTIONS. 8.1 Random Sampling. 8.2 Some Important Statistics CHAPTER 8 FUNDAMENTAL SAMPLING DISTRIBUTIONS AND DATA DESCRIPTIONS 8.1 Radom Samplig The basic idea of the statistical iferece is that we are allowed to draw ifereces or coclusios about a populatio based

More information

Chapter 8: Estimating with Confidence

Chapter 8: Estimating with Confidence Chapter 8: Estimatig with Cofidece Sectio 8.2 The Practice of Statistics, 4 th editio For AP* STARNES, YATES, MOORE Chapter 8 Estimatig with Cofidece 8.1 Cofidece Itervals: The Basics 8.2 8.3 Estimatig

More information

AP Statistics Review Ch. 8

AP Statistics Review Ch. 8 AP Statistics Review Ch. 8 Name 1. Each figure below displays the samplig distributio of a statistic used to estimate a parameter. The true value of the populatio parameter is marked o each samplig distributio.

More information

The Random Walk For Dummies

The Random Walk For Dummies The Radom Walk For Dummies Richard A Mote Abstract We look at the priciples goverig the oe-dimesioal discrete radom walk First we review five basic cocepts of probability theory The we cosider the Beroulli

More information

10. Comparative Tests among Spatial Regression Models. Here we revisit the example in Section 8.1 of estimating the mean of a normal random

10. Comparative Tests among Spatial Regression Models. Here we revisit the example in Section 8.1 of estimating the mean of a normal random Part III. Areal Data Aalysis 0. Comparative Tests amog Spatial Regressio Models While the otio of relative likelihood values for differet models is somewhat difficult to iterpret directly (as metioed above),

More information

6.3 Testing Series With Positive Terms

6.3 Testing Series With Positive Terms 6.3. TESTING SERIES WITH POSITIVE TERMS 307 6.3 Testig Series With Positive Terms 6.3. Review of what is kow up to ow I theory, testig a series a i for covergece amouts to fidig the i= sequece of partial

More information

A NEW METHOD FOR CONSTRUCTING APPROXIMATE CONFIDENCE INTERVALS FOR M-ESTU1ATES. Dennis D. Boos

A NEW METHOD FOR CONSTRUCTING APPROXIMATE CONFIDENCE INTERVALS FOR M-ESTU1ATES. Dennis D. Boos .- A NEW METHOD FOR CONSTRUCTING APPROXIMATE CONFIDENCE INTERVALS FOR M-ESTU1ATES by Deis D. Boos Departmet of Statistics North Carolia State Uiversity Istitute of Statistics Mimeo Series #1198 September,

More information

There is no straightforward approach for choosing the warmup period l.

There is no straightforward approach for choosing the warmup period l. B. Maddah INDE 504 Discrete-Evet Simulatio Output Aalysis () Statistical Aalysis for Steady-State Parameters I a otermiatig simulatio, the iterest is i estimatig the log ru steady state measures of performace.

More information

KLMED8004 Medical statistics. Part I, autumn Estimation. We have previously learned: Population and sample. New questions

KLMED8004 Medical statistics. Part I, autumn Estimation. We have previously learned: Population and sample. New questions We have previously leared: KLMED8004 Medical statistics Part I, autum 00 How kow probability distributios (e.g. biomial distributio, ormal distributio) with kow populatio parameters (mea, variace) ca give

More information

Binomial Distribution

Binomial Distribution 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 1 2 3 4 5 6 7 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Overview Example: coi tossed three times Defiitio Formula Recall that a r.v. is discrete if there are either a fiite umber of possible

More information

Chapter 11: Asking and Answering Questions About the Difference of Two Proportions

Chapter 11: Asking and Answering Questions About the Difference of Two Proportions Chapter 11: Askig ad Aswerig Questios About the Differece of Two Proportios These otes reflect material from our text, Statistics, Learig from Data, First Editio, by Roxy Peck, published by CENGAGE Learig,

More information

Expectation and Variance of a random variable

Expectation and Variance of a random variable Chapter 11 Expectatio ad Variace of a radom variable The aim of this lecture is to defie ad itroduce mathematical Expectatio ad variace of a fuctio of discrete & cotiuous radom variables ad the distributio

More information

Investigating the Significance of a Correlation Coefficient using Jackknife Estimates

Investigating the Significance of a Correlation Coefficient using Jackknife Estimates Iteratioal Joural of Scieces: Basic ad Applied Research (IJSBAR) ISSN 2307-4531 (Prit & Olie) http://gssrr.org/idex.php?joural=jouralofbasicadapplied ---------------------------------------------------------------------------------------------------------------------------

More information

Output Analysis and Run-Length Control

Output Analysis and Run-Length Control IEOR E4703: Mote Carlo Simulatio Columbia Uiversity c 2017 by Marti Haugh Output Aalysis ad Ru-Legth Cotrol I these otes we describe how the Cetral Limit Theorem ca be used to costruct approximate (1 α%

More information

Varanasi , India. Corresponding author

Varanasi , India. Corresponding author A Geeral Family of Estimators for Estimatig Populatio Mea i Systematic Samplig Usig Auxiliary Iformatio i the Presece of Missig Observatios Maoj K. Chaudhary, Sachi Malik, Jayat Sigh ad Rajesh Sigh Departmet

More information

2 1. The r.s., of size n2, from population 2 will be. 2 and 2. 2) The two populations are independent. This implies that all of the n1 n2

2 1. The r.s., of size n2, from population 2 will be. 2 and 2. 2) The two populations are independent. This implies that all of the n1 n2 Chapter 8 Comparig Two Treatmets Iferece about Two Populatio Meas We wat to compare the meas of two populatios to see whether they differ. There are two situatios to cosider, as show i the followig examples:

More information

Introducing Sample Proportions

Introducing Sample Proportions Itroducig Sample Proportios Probability ad statistics Aswers & Notes TI-Nspire Ivestigatio Studet 60 mi 7 8 9 0 Itroductio A 00 survey of attitudes to climate chage, coducted i Australia by the CSIRO,

More information

OPTIMAL ALGORITHMS -- SUPPLEMENTAL NOTES

OPTIMAL ALGORITHMS -- SUPPLEMENTAL NOTES OPTIMAL ALGORITHMS -- SUPPLEMENTAL NOTES Peter M. Maurer Why Hashig is θ(). As i biary search, hashig assumes that keys are stored i a array which is idexed by a iteger. However, hashig attempts to bypass

More information

Chapter 22. Comparing Two Proportions. Copyright 2010, 2007, 2004 Pearson Education, Inc.

Chapter 22. Comparing Two Proportions. Copyright 2010, 2007, 2004 Pearson Education, Inc. Chapter 22 Comparig Two Proportios Copyright 2010, 2007, 2004 Pearso Educatio, Ic. Comparig Two Proportios Read the first two paragraphs of pg 504. Comparisos betwee two percetages are much more commo

More information

The standard deviation of the mean

The standard deviation of the mean Physics 6C Fall 20 The stadard deviatio of the mea These otes provide some clarificatio o the distictio betwee the stadard deviatio ad the stadard deviatio of the mea.. The sample mea ad variace Cosider

More information

Abstract. Ranked set sampling, auxiliary variable, variance.

Abstract. Ranked set sampling, auxiliary variable, variance. Hacettepe Joural of Mathematics ad Statistics Volume (), 1 A class of Hartley-Ross type Ubiased estimators for Populatio Mea usig Raked Set Samplig Lakhkar Kha ad Javid Shabbir Abstract I this paper, we

More information

Chapter 6 Sampling Distributions

Chapter 6 Sampling Distributions Chapter 6 Samplig Distributios 1 I most experimets, we have more tha oe measuremet for ay give variable, each measuremet beig associated with oe radomly selected a member of a populatio. Hece we eed to

More information

Basis for simulation techniques

Basis for simulation techniques Basis for simulatio techiques M. Veeraraghava, March 7, 004 Estimatio is based o a collectio of experimetal outcomes, x, x,, x, where each experimetal outcome is a value of a radom variable. x i. Defiitios

More information

Chapter 22. Comparing Two Proportions. Copyright 2010 Pearson Education, Inc.

Chapter 22. Comparing Two Proportions. Copyright 2010 Pearson Education, Inc. Chapter 22 Comparig Two Proportios Copyright 2010 Pearso Educatio, Ic. Comparig Two Proportios Comparisos betwee two percetages are much more commo tha questios about isolated percetages. Ad they are more

More information

1.010 Uncertainty in Engineering Fall 2008

1.010 Uncertainty in Engineering Fall 2008 MIT OpeCourseWare http://ocw.mit.edu.00 Ucertaity i Egieerig Fall 2008 For iformatio about citig these materials or our Terms of Use, visit: http://ocw.mit.edu.terms. .00 - Brief Notes # 9 Poit ad Iterval

More information

GUIDELINES ON REPRESENTATIVE SAMPLING

GUIDELINES ON REPRESENTATIVE SAMPLING DRUGS WORKING GROUP VALIDATION OF THE GUIDELINES ON REPRESENTATIVE SAMPLING DOCUMENT TYPE : REF. CODE: ISSUE NO: ISSUE DATE: VALIDATION REPORT DWG-SGL-001 002 08 DECEMBER 2012 Ref code: DWG-SGL-001 Issue

More information

Chapter 6 Principles of Data Reduction

Chapter 6 Principles of Data Reduction Chapter 6 for BST 695: Special Topics i Statistical Theory. Kui Zhag, 0 Chapter 6 Priciples of Data Reductio Sectio 6. Itroductio Goal: To summarize or reduce the data X, X,, X to get iformatio about a

More information

ENGI 4421 Confidence Intervals (Two Samples) Page 12-01

ENGI 4421 Confidence Intervals (Two Samples) Page 12-01 ENGI 44 Cofidece Itervals (Two Samples) Page -0 Two Sample Cofidece Iterval for a Differece i Populatio Meas [Navidi sectios 5.4-5.7; Devore chapter 9] From the cetral limit theorem, we kow that, for sufficietly

More information

Product measures, Tonelli s and Fubini s theorems For use in MAT3400/4400, autumn 2014 Nadia S. Larsen. Version of 13 October 2014.

Product measures, Tonelli s and Fubini s theorems For use in MAT3400/4400, autumn 2014 Nadia S. Larsen. Version of 13 October 2014. Product measures, Toelli s ad Fubii s theorems For use i MAT3400/4400, autum 2014 Nadia S. Larse Versio of 13 October 2014. 1. Costructio of the product measure The purpose of these otes is to preset the

More information

Hypothesis Testing. Evaluation of Performance of Learned h. Issues. Trade-off Between Bias and Variance

Hypothesis Testing. Evaluation of Performance of Learned h. Issues. Trade-off Between Bias and Variance Hypothesis Testig Empirically evaluatig accuracy of hypotheses: importat activity i ML. Three questios: Give observed accuracy over a sample set, how well does this estimate apply over additioal samples?

More information

An Introduction to Randomized Algorithms

An Introduction to Randomized Algorithms A Itroductio to Radomized Algorithms The focus of this lecture is to study a radomized algorithm for quick sort, aalyze it usig probabilistic recurrece relatios, ad also provide more geeral tools for aalysis

More information

Infinite Sequences and Series

Infinite Sequences and Series Chapter 6 Ifiite Sequeces ad Series 6.1 Ifiite Sequeces 6.1.1 Elemetary Cocepts Simply speakig, a sequece is a ordered list of umbers writte: {a 1, a 2, a 3,...a, a +1,...} where the elemets a i represet

More information

STA Learning Objectives. Population Proportions. Module 10 Comparing Two Proportions. Upon completing this module, you should be able to:

STA Learning Objectives. Population Proportions. Module 10 Comparing Two Proportions. Upon completing this module, you should be able to: STA 2023 Module 10 Comparig Two Proportios Learig Objectives Upo completig this module, you should be able to: 1. Perform large-sample ifereces (hypothesis test ad cofidece itervals) to compare two populatio

More information

Some Exponential Ratio-Product Type Estimators using information on Auxiliary Attributes under Second Order Approximation

Some Exponential Ratio-Product Type Estimators using information on Auxiliary Attributes under Second Order Approximation ; [Formerly kow as the Bulleti of Statistics & Ecoomics (ISSN 097-70)]; ISSN 0975-556X; Year: 0, Volume:, Issue Number: ; It. j. stat. eco.; opyright 0 by ESER Publicatios Some Expoetial Ratio-Product

More information

Economics Spring 2015

Economics Spring 2015 1 Ecoomics 400 -- Sprig 015 /17/015 pp. 30-38; Ch. 7.1.4-7. New Stata Assigmet ad ew MyStatlab assigmet, both due Feb 4th Midterm Exam Thursday Feb 6th, Chapters 1-7 of Groeber text ad all relevat lectures

More information

Bootstrap Intervals of the Parameters of Lognormal Distribution Using Power Rule Model and Accelerated Life Tests

Bootstrap Intervals of the Parameters of Lognormal Distribution Using Power Rule Model and Accelerated Life Tests Joural of Moder Applied Statistical Methods Volume 5 Issue Article --5 Bootstrap Itervals of the Parameters of Logormal Distributio Usig Power Rule Model ad Accelerated Life Tests Mohammed Al-Ha Ebrahem

More information

Continuous Data that can take on any real number (time/length) based on sample data. Categorical data can only be named or categorised

Continuous Data that can take on any real number (time/length) based on sample data. Categorical data can only be named or categorised Questio 1. (Topics 1-3) A populatio cosists of all the members of a group about which you wat to draw a coclusio (Greek letters (μ, σ, Ν) are used) A sample is the portio of the populatio selected for

More information

Confidence Intervals

Confidence Intervals Cofidece Itervals Berli Che Deartmet of Comuter Sciece & Iformatio Egieerig Natioal Taiwa Normal Uiversity Referece: 1. W. Navidi. Statistics for Egieerig ad Scietists. Chater 5 & Teachig Material Itroductio

More information

April 18, 2017 CONFIDENCE INTERVALS AND HYPOTHESIS TESTING, UNDERGRADUATE MATH 526 STYLE

April 18, 2017 CONFIDENCE INTERVALS AND HYPOTHESIS TESTING, UNDERGRADUATE MATH 526 STYLE April 18, 2017 CONFIDENCE INTERVALS AND HYPOTHESIS TESTING, UNDERGRADUATE MATH 526 STYLE TERRY SOO Abstract These otes are adapted from whe I taught Math 526 ad meat to give a quick itroductio to cofidece

More information

Estimation for Complete Data

Estimation for Complete Data Estimatio for Complete Data complete data: there is o loss of iformatio durig study. complete idividual complete data= grouped data A complete idividual data is the oe i which the complete iformatio of

More information

17. Joint distributions of extreme order statistics Lehmann 5.1; Ferguson 15

17. Joint distributions of extreme order statistics Lehmann 5.1; Ferguson 15 17. Joit distributios of extreme order statistics Lehma 5.1; Ferguso 15 I Example 10., we derived the asymptotic distributio of the maximum from a radom sample from a uiform distributio. We did this usig

More information

UNIT 8: INTRODUCTION TO INTERVAL ESTIMATION

UNIT 8: INTRODUCTION TO INTERVAL ESTIMATION STATISTICAL METHODS FOR BUSINESS UNIT 8: INTRODUCTION TO INTERVAL ESTIMATION 8..- Itroductio to iterval estimatio 8..- Cofidece itervals. Costructio ad characteristics 8.3.- Cofidece itervals for the mea

More information

Statistical Intervals for a Single Sample

Statistical Intervals for a Single Sample 3/5/06 Applied Statistics ad Probability for Egieers Sixth Editio Douglas C. Motgomery George C. Ruger Chapter 8 Statistical Itervals for a Sigle Sample 8 CHAPTER OUTLINE 8- Cofidece Iterval o the Mea

More information

MOST PEOPLE WOULD RATHER LIVE WITH A PROBLEM THEY CAN'T SOLVE, THAN ACCEPT A SOLUTION THEY CAN'T UNDERSTAND.

MOST PEOPLE WOULD RATHER LIVE WITH A PROBLEM THEY CAN'T SOLVE, THAN ACCEPT A SOLUTION THEY CAN'T UNDERSTAND. XI-1 (1074) MOST PEOPLE WOULD RATHER LIVE WITH A PROBLEM THEY CAN'T SOLVE, THAN ACCEPT A SOLUTION THEY CAN'T UNDERSTAND. R. E. D. WOOLSEY AND H. S. SWANSON XI-2 (1075) STATISTICAL DECISION MAKING Advaced

More information

1 Introduction to reducing variance in Monte Carlo simulations

1 Introduction to reducing variance in Monte Carlo simulations Copyright c 010 by Karl Sigma 1 Itroductio to reducig variace i Mote Carlo simulatios 11 Review of cofidece itervals for estimatig a mea I statistics, we estimate a ukow mea µ = E(X) of a distributio by

More information

Statisticians use the word population to refer the total number of (potential) observations under consideration

Statisticians use the word population to refer the total number of (potential) observations under consideration 6 Samplig Distributios Statisticias use the word populatio to refer the total umber of (potetial) observatios uder cosideratio The populatio is just the set of all possible outcomes i our sample space

More information

MATH/STAT 352: Lecture 15

MATH/STAT 352: Lecture 15 MATH/STAT 352: Lecture 15 Sectios 5.2 ad 5.3. Large sample CI for a proportio ad small sample CI for a mea. 1 5.2: Cofidece Iterval for a Proportio Estimatig proportio of successes i a biomial experimet

More information

Overview. p 2. Chapter 9. Pooled Estimate of. q = 1 p. Notation for Two Proportions. Inferences about Two Proportions. Assumptions

Overview. p 2. Chapter 9. Pooled Estimate of. q = 1 p. Notation for Two Proportions. Inferences about Two Proportions. Assumptions Chapter 9 Slide Ifereces from Two Samples 9- Overview 9- Ifereces about Two Proportios 9- Ifereces about Two Meas: Idepedet Samples 9-4 Ifereces about Matched Pairs 9-5 Comparig Variatio i Two Samples

More information

The variance of a sum of independent variables is the sum of their variances, since covariances are zero. Therefore. V (xi )= n n 2 σ2 = σ2.

The variance of a sum of independent variables is the sum of their variances, since covariances are zero. Therefore. V (xi )= n n 2 σ2 = σ2. SAMPLE STATISTICS A radom sample x 1,x,,x from a distributio f(x) is a set of idepedetly ad idetically variables with x i f(x) for all i Their joit pdf is f(x 1,x,,x )=f(x 1 )f(x ) f(x )= f(x i ) The sample

More information