Small area estimation under spatial SAR model

Size: px
Start display at page:

Download "Small area estimation under spatial SAR model"

Transcription

1 Small area estimation under spatial SAR model Jan Kubacki Centre for Mathematical Statistics, Statistical Office in Łódź Alina Jędrzejczak Chair of Statistical Methods, Institute of Econometrics and Statistics, University of Łódź, Poland

2 Summary In the presentation the method of small area estimation under spatial Simultaneous Autoregressive (SAR) model is presented. The estimation was conducted using both spatial EBLUP and hierarchical Bayes method with SAR random effects that depend on proximity matrix and spatial autoregression coefficient (ρ). The presentation continues the idea of small area estimation by means of hierarchical Bayes method in the case of known model hyperparameters, presented in a previous work of one of the authors. The computations performed by sae package for R-project environment and special procedure prepared for WinBUGS software reveal that consistent estimates (for point estimates, for measures of variability - such as estimation error and for random effects) can be obtained. As an example the data on average per capita available income from Polish Household Budget Survey for counties (NUTS4) and auxiliary variables from Polish Tax Register (POLTAX) were used together with digital maps for Polish counties. The precision estimation for direct estimates were obtained by means of three-stage method, where Balanced Repeated Replication (BRR), bootstrap and Generalized Variance Function were used. The results of the computations reveal that for higher values of ρ some MSE reduction can be observed. This effect for HB-SAR method is even more evident than for common spatial EBLUP. The computations done with Gibbs sampler also demonstrate simultaneous nature of simulated processes especially for random effects. In our opinion, the Gibbs sampler can be a good starting point for other computation methods that seems more adequate for this kind of stochastic processes.

3 Mixed model for small areas In this work we discuss a special case of area level model (type A model), where the parameter of interest is vector θ of size m (where m is the number of small areas), which is related to the direct estimator θˆ with the following relationship Where e is the vector of independent sampling errors having mean 0 and diagonal variance matrix ψ. Parameter θ satisfies the following relationships which incorporates the spatial relationships between areas θ θˆ θ Xβ where X is the matrix of area dependent auxiliary variables of size m p, β is the regression parameters vector of size p, Z is the matrix (m m size) of known positive constants and v is the vector of independent and identically distributed variables (called random effects) with mean 0 and variance σ v e Zv () ()

4 SAR model specification There are two common variants of spatial models depending on the random effects pattern. The first one is the simultaneously autoregressive model (SAR) and the second is the conditional autoregressive model (CAR). In our presentation we assume that the random effects are described by SAR process with random effects described as follows v Wv W) Where ρ is the spatial autoregressive coefficient, W is the proximity matrix (of size m m), u is the vector of independent error term with zero mean and constant variance σ u and I is the identity matrix of size m m. The random effects have the following covariance matrix G (called SAR dispersion matrix) G and the sampling error e has the following covariance matrix u u v ( I T [( I W) ( I W )] u (3) R Ψ diag ( i )

5 Spatial EBLUP estimator When the matrix Z is neglected (or equal I) the model () with spatially-correlated random effects takes the form: θˆ Xβ ( I W ) u e Under the above model the Spatial EBLUP estimator is equal to ~ θ i x T i ~ β b T i GV (ˆ θ ~ Xβ) ~ T T Where β X V X X V θˆ is the generalized LS estimator of the i [ ] regression parameter and b T i is the m vector (0,,0,,0, 0) with in the i-th position. This estimator is dependent on the σ u and ρ which can be estimated using ML or REML method.

6 MSE for spatial EBLUP estimator The mean square error for Spatial EBLUP estimator can be expressed as the sum of four following elements ~ mse[ i ] g g g3 g 4 where g is connected with uncertainty about the small area estimate and is of order O(), g is connected with uncertainty about β and is of order O(m - ) for large m, g 3 is connected with uncertainty about σ u (or variance components) and g 4 is connected with uncertainty of spatial autocorrelation parameter ρ. The first two elements are given by gi T b i [ G GV G] b i g i b T i T T [ I m GV ] X(X V X) X [ I m V G ] b i The third element is more complicated and can be expressed as g i 3 trace{ L VL T i i I } where L i b T i b [ C T i V [ AV u u C C V V C AV V ] ] C ( I W) ( I W) m A C ( W W W W ) C T T T m u and I - is the Fisher information matrix inverse

7 Estimation of g 4 element for MSE The last term g 4 can be expressed by where the second derivatives are given by the following formulas: i b Ψ V ΨV b ) ( ) ( 4 kl l k k l T i i I V g m m u V 0 ) ( ) ( ) ( ) ( C C C V V u u T WC W C C C C C C V ) ( u u

8 Hierarchical model incorporating SAR process In the case where ρ and σ u are known (could be obtained from Spatial EBLUP estimation), a spatial hierarchical Bayes model can be specified. Its form is similar to the usual HB model (Rao, 003, eq. (0.3.)), but it accounts for spatial dependences between small areas. The model can be expressed as follows (i) (ii) (iii) (iv) θ β, ˆ θ, β, u θ f (β) u u, ~ β, θ, θˆ N( Xβ, ~ ~ u N( θ, Ψ) G - [( I ( a, b) m W ) W )] The model assumes known prior Gamma distribution parameters for σ u. In our case they were obtained from previous studies (EBLUP estimators achieved for 6 regions and two years what may be a good approximation of σ u variability). Similar approach was used in Kubacki (0), where the distribution of σ u was obtained from ordinary regression models. T ( I m )

9 Empirical distribution of model variance obtained for EBLUP estimators

10 WinBUGS computational models Conventional model for known σ u model { for(p in : N) { Y[p] ~ dnorm(mu[p], tau[p]) mu[p] <- alpha[] + alpha[] * A[p] + alpha[3] * B[p] + u[p] u[p] ~ dnorm(0, precu) } precu ~ dgamma (a0,b0) alpha[] ~ dflat() alpha[] ~ dflat() alpha[3] ~ dflat() sigmau<-/precu } Model for SAR version model { for(p in : N) { Y[p] ~ dnorm(mu[p], tau[p]) mu[p] <- alpha[] + alpha[] * A[p] + alpha[3] * B[p] + v[p] u[p] ~ dnorm(0, precu) v[p] <- inprod(rho_w[p,:n],u[:n]) } precu ~ dgamma (a0,b0) alpha[] ~ dflat() alpha[] ~ dflat() alpha[3] ~ dflat() sigmau<-/precu }

11 R-project macro determining the model parameters # fitting the Spatial EBLUP model resultsreml <- eblupsfh(direst ~ + A + B, desvar, W, method="reml", data=d, MAXITER=500) msesreml <- msesfh(direst ~ + A + B, desvar, W, method="reml", data=d, MAXITER=500) sigmas_reml <- resultsreml$fit$refvar rho_reml <- resultsreml$fit$spatialcorr # determining the model parameters I <-diag(,n) for (j in :N) { W_row <- W[j,] for (k in :lpow) { W_mat[j,k] <- W_row[k] } } rho_w <- solve(i-rho_reml*w_mat) a0 <- dochg_shape_sp b0 <- dochg_rate_sp infile <- "coda.txt" indfile <- "codaindex.txt" data <- list(n=n, Y=Y, tau=tau, A=A, B=B, a0=a0, b0=b0, rho_w=rho_w) model <- lm( ydir[,] ~ + A + B) mod_smry <- summary(model) alpha <- as.vector(mod_smry$coefficients[,]) sigma_ <- (mod_smry$sigma)*(mod_smry$sigma) precu <- /sigma_

12 R-project macro making the simulations v <- vector(mode = "numeric", length = N) u <- vector(mode = "numeric", length = N) inits <- list(list(alpha=alpha, precu=precu, u=u)) parameters <- c("mu", "alpha", "precu", "v", "u") working.directory <- getwd() # simulations - WinBUGS call and collecting the data sim_hb <- bugs(data, inits, parameters, model_hb,n.chains=, n.burnin =, n.iter=0000, n.thin =, codapkg=true, working.directory = working.directory) results <- read.coda(infile, indfile,, 0000, )

13 Available income estimates with their precision and random effect estimates-part County (NUTS 4 unit) Direct estimates Parameter estimate REE % Available income EBLUP estimates REML Parameter estimate REE % Random effects HB estimates Parameter estimate REE % Random effects będziński 8,59 5,8 784,7 4,8 3,8 789,5 4,40 7,5 bielski 78,94, 778,63,0 53,9 779,40,9 54,8 cieszyński 76,78 4,99 734, 3,9 9, 738,6 3,97 33,7 częstochowski 570,65 6,54 590,44 5,00 -,00 586,37 5,08 -,73 gliwicki 693,0,4 706,85 5,9-3,38 705,97 5,6-3,95 kłobucki 539,04 8,00 574,9 5,59-8,0 568,43 5,77-30,7 lubliniecki 596,73 4,4 60,8 3,63-34, 608,03 3,63-34,85 mikołowski 796,64 5,83 793,96 5,0 0,5 796,39 5,4-0,06 myszkowski 63,46 7,74 6,49 5,35-5,9 60,0 5,4-5,43 pszczyński 69,75,4 74,5 5,4-3,85 79,7 5,79-30, raciborski 758,34 4,50 76,66 3,88 5,76 7,3 3,94 59,80 rybnicki 783,98 8,8 764,5 4,69 7, 766,0 5,08 7,97 tarnogórski 67,46 5,07 686,73 3,9-9,45 684,87 3,99 -,04 bieruńsko-lędziński 65,85,69 764,96 4,93-38,35 757, 5,3-49,9 wodzisławski 855,7 4,4 8,73 3,53 48,34 89,80 3,66 54,3 zawierciański 67,04 7,36 674,0 4,89 -,80 67,0 5,06 -,8 żywiecki 730,75,80 75,43,75 45,63 76,56,75 47,68 m.bielsko-biała 79,4 7,38 77,68 4,4 8,84 774,7 4,66 9,78

14 Available income estimates with their precision and random effect estimates-part County (NUTS 4 unit) Direct estimates Parameter estimate REE % Available income EBLUP estimates REML Parameter estimate REE % Random effects HB estimates Parameter estimate REE % Random effects m.bytom 705,56,9 705,8,7 4,93 705,3,6 5,59 m.chorzów 656,8,50 666,95,34-58,78 664,70,37-6,08 m.częstochowa 77,36 0,35 75,50 5,,94 79,7 5,57 6,93 m.dąbrowa Górnicza 777,5 4,38 78,6 3,47-6,48 783,03 3,47-8,58 m.gliwice 745,60 5,50 76,7 3,9-3,66 760,5 3,96-6,57 m.jastrzębie-zdrój 748,66 5,5 774,86 3,90 -,65 77,66 4,0-8,39 m.jaworzno 748,49 6,85 780, 4, -7,74 778,39 4,4 -, m.katowice 859,53,3 859,7, 5,66 859,38,,5 m.mysłowice 83,9,4 80,39,3 0,79 8,09,0 8,73 m.piekary Śląskie 744,4 3,4 74,89 5,3 0,35 74,68 5,88-0,4 m.ruda Śląska 67,9 5,46 77,90 5,09-3,8 768,99 5,75-9,80 m.rybnik 763,03,45 764,5,4-7,86 764,0,4-0,39 m.siemianowice Śląskie 95,69 9,5 774,56 4,93 9,68 783,96 5,63 38,39 m.sosnowiec 88,88 7,44 803,93 4,39 5,95 806,06 4,65 5,59 m.świętochłowice 686,8 9,03 708,8 4,90-8,4 706,4 5,8-0,05 m.tychy 83,03 0,07 83,03 0,07-4,80 83,0 0,07-9,04 m.zabrze 703,7 0, 703,74 0, -5,66 703,73 0, -5,83 m.żory 830,68 8,0 78,97 4,53 5,50 786,79 4,88 8,58

15 Observed vs. predicted available income for HB estimation Direct estimator (double black circles), EBLUP estimator (red circles)

16 Plot of random effects for EBLUP vs. HB method

17 Plots of distributions of model estimates for available income obtained by MCMC simulation using conventional model

18 Plots of autocorrelations of model estimates for available income per capita obtained for Silesian region by MCMC simulation using Gibbs sampler with conventional model

19 Available income estimates with their precision and random effect estimates-part County (NUTS 4 unit) Direct estimates Parameter estimate REE % Available income EBLUP estimates Spatial REML Parameter estimate REE % Random effects HB SAR estimates Parameter estimate REE % Random effects będziński 8,59 5,8 78,77 3,95,06 785,6 3,87,85 bielski 78,94, 779,45,0 59,60 780,,6 67,8 cieszyński 76,78 4,99 746,3 3,5 4,66 748,93 3,5 5,75 częstochowski 570,65 6,54 584,70 4,74-36,0 58, 4,68-3,99 gliwicki 693,0,4 7,7 4,55-5, 74,97 3,88 5,50 kłobucki 539,04 8,00 569,85 5,37-4,6 566,06 5,40-38,9 lubliniecki 596,73 4,4 603,87 3,58-4,7 60,7 3,54-37,00 mikołowski 796,64 5,83 788,05 4,4 4, 798,48 5,5,67 myszkowski 63,46 7,74 68,68 4,95 -,87 67,30 4,89-7,47 pszczyński 69,75,4 734,98 4,50-0,3 747,9 4,78 0,4 raciborski 758,34 4,50 70,69 4,00 49, 78,45 3,90 63,98 rybnicki 783,98 8,8 780,3 4,0 7,6 78, 4,34 6,94 tarnogórski 67,46 5,07 683,65 3,66-4,50 684,5 3,56-6,3 bieruńsko-lędziński 65,85,69 770,56 4,07-0,5 776,08 4,08-6,79 wodzisławski 855,7 4,4 83,85 3,47 4,0 89,97 3,56 56,09 zawierciański 67,04 7,36 673,66 4,56 -,07 67,67 4,64-6,88 żywiecki 730,75,80 79,35,75 5,79 79,74,69 59,34 m.bielsko-biała 79,4 7,38 799,78 3,96 4,89 797,44 4,03 47,44

20 Available income estimates with their precision and random effect estimates-part County (NUTS 4 unit) Direct estimates Parameter estimate REE % Available income EBLUP estimates Spatial REML Parameter estimate REE % Random effects HB SAR estimates Parameter estimate REE % Random effects m.bytom 705,56,9 703,46,7-6,94 703,80,6 0,8 m.chorzów 656,8,50 67,64,3-6,83 668,96,8-58,88 m.częstochowa 77,36 0,35 68,88 5,3-9,93 683,66 5,4 -,78 m.dąbrowa Górnicza 777,5 4,38 78,40 3,48-0,4 780,4 3,35 6,7 m.gliwice 745,60 5,50 753,9 3,78-7,6 753,8 3,57-9,8 m.jastrzębie-zdrój 748,66 5,5 795, 3,67 -,68 789,5 3,6 0,94 m.jaworzno 748,49 6,85 784,3 3,88-0,3 78,63 3,9-3,53 m.katowice 859,53,3 857,99, 9,63 859,,07 9,47 m.mysłowice 83,9,4 806,30,,49 806,86,5 0,7 m.piekary Śląskie 744,4 3,4 738,50 4,50 -,0 738,6 4,86-4,48 m.ruda Śląska 67,9 5,46 765,34 4,9-5,3 758,58 4,60-3,85 m.rybnik 763,03,45 768,,4-0, 767,0,39-3, m.siemianowice Śląskie 95,69 9,5 760,5 4,35 7,50 768,9 5,00 3,3 m.sosnowiec 88,88 7,44 807, 3,85 8,00 805,53 4,00 4,55 m.świętochłowice 686,8 9,03 69,04 4,5-9,30 689,37 4,76-4,46 m.tychy 83,03 0,07 83,03 0,07 0,70 83,0 0,07 9,8 m.zabrze 703,7 0, 703,73 0, -6,69 703,7 0, -9,8 m.żory 830,68 8,0 770,40 4,9 8,08 773,90 4,45 9,50

21 Observed vs. predicted available income for HB SAR estimation, Direct estimators (double black circles), Spatial EBLUP estimators (red circles)

22 Plot of random effects for Spatial EBLUP vs. HB-SAR method

23 Plots of distributions of model estimates for available income obtained by MCMC simulation using SAR model

24 Plots of autocorrelations of model estimates for available income per capita obtained for Silesian region by MCMC with SAR relationships simulation using Gibbs sampler

25 Relative Estimation Error Reduction and Spatial Gain for Silesian region obtained using EBLUP and HB method with Gibbs sampler County (NUTS-4 unit) Relative estimation error reduction EBLUP HB Spatial EBLUP HB-SAR Spatial estimation error reduction Spatial EBLUP HB-SAR będziński,36,3,476,504,084,37 bielski,03,06,05,049,00,03 cieszyński,73,57,4,40,7,30 częstochowski,307,86,379,396,055,086 gliwicki,45,987,448,869,4,444 kłobucki,430,386,488,48,04,069 lubliniecki,40,38,56,67,03,05 mikołowski 3,59,94 3,737 3,074,83,05 myszkowski,446,49,563,583,08,08 pszczyński,4,097,698,54,03, raciborski,6,44,5,55 0,969,00 rybnicki,746,60,950,886,6,7 tarnogórski,94,7,385,43,07,8 bieruńsko-lędziński,374,99,87,867,09,303 wodzisławski,99,56,,90,09,09 zawierciański,506,454,64,588,07,09 żywiecki,04,04,08,065,004,040 m.bielsko-biała,669,58,864,89,7,56

26 Relative Estimation Error reduction and Spatial Gain for estimation error for Silesian region obtained using EBLUP and HB method with Gibbs sampler County (NUTS-4 unit) Relative estimation error reduction EBLUP HB Spatial EBLUP HB-SAR Spatial estimation error reduction Spatial EBLUP HB-SAR m.bytom,04,0,0,09 0,997 0,998 m.chorzów,066,05,08,093,04,040 m.częstochowa,0,858,949,908 0,964,07 m.dąbrowa Górnicza,63,6,6,306 0,999,035 m.gliwice,407,388,456,543,035, m.jastrzębie-zdrój,47,375,504,53,06,3 m.jaworzno,64,550,767,749,088,8 m.katowice,04,04,00,053 0,996,039 m.mysłowice,077,095,085,7,007,00 m.piekary Śląskie,473,34,9,706,8, m.ruda Śląska 3,035,689 3,60 3,359,87,50 m.rybnik,07,09,03,046,005,07 m.siemianowice Śląskie,857,66,03,83,3,6 m.sosnowiec,693,598,93,858,4,6 m.świętochłowice,844,70,000,897,085,09 m.tychy,000,0,000 0,994,000 0,983 m.zabrze,000 0,99,000 0,99,000,00 m.żory,8,68,959,845,08,097

27 Absolute values of random effects for available per capita income (Silesian region )

28 Distribution of relative estimation error for: Direct estimator, EBLUP (spatial and ordinary - REML variant) and HB (both ordinary and SAR)

29 Distribution of REE reduction for EBLUP (spatial and ordinary - REML variant) and HB (both ordinary and SAR) estimators

30 Distribution of REE reduction due to spatial relationships for Spatial EBLUP and HB-SAR estimators

31 REE reduction for Spatial EBLUP and HB-SAR estimators obtained for different a priori spatial autoregressive coefficient values (simulation study)

32 REE reduction due to spatial relationships for Spatial EBLUP and HB-SAR estimators for different a priori spatial autoregressive coefficient values

33 Trace of results for Gibbs sampler simulation obtained for first 3 values of random effects (u description) obtained for ordinary HB method and first 3 values of random effects (v description) obtained for HB-SAR method for 004 year and Silesian region, assuming (for v values) that the spatial autoregressive coefficient is equal 0.95

34 Conclusions The results presented in the presentation show that, using the auxiliary information on spatial neighborhood, one can obtain relatively precise estimates based on small area hierarchical models. In particular, it is possible by means of spatial Simultaneous Autoregressive Process (SAR). The advantage of Spatial HB estimation observed in terms of a realworld example prepared for Silesian region and the data from Polish HBS and Polish tax register may not be so clear for other regions. The simulation-based calculations also confirmed noticeable gains in efficiency for Spatial HB estimators, when some assumptions on spatial autoregressive coefficient have been made. This effect, however, is more evident for Spatial HB method than for Spatial EBLUP technique.

35 Literature Cressie, N. (99). Small-area prediction of undercount using the general linear model. Proceedings of Statistic Symposium 90: Measurement and Improvement of Data Quality, Ottawa: Statistics Canada, Gharde, Y., Rai, A., Jaggi S. (03), Bayesian Prediction in Spatial Small Area Models, Journal of the Indian Society of Agricultural Statisitcs, 67(3), Gilks, W.R., Best, N.G. and Tan, K.K.C. (995). Adaptive Rejection Metropolis Sampling Within Gibbs Sampling. Applied Statistics, 44, Gomez-Rubio, V., (008), Small Area Estimation with R Unit 5: Bayesian Small Area Estimation, user!008 August 008, Dortmund (Germany) Kubacki, J., (0) Estimation of parameters for small areas using hierarchical Bayes method in the case of known model hyperparameters, Statistics in Transition-new series, Summer 0, Vol. 3, No., 6 78 Kubacki, J., Jędrzejczak A., (0) The Comparison of Generalized Variance Function with Other Methods of Precision Estimation for Polish Household Budget Survey, Studia Ekonomiczne, 0, Molina I., Marhuenda Y., (03) Package sae, Functions for small area estimation, De Oliveira, V., Jin Song, J., (008), Bayesian Analysis of Simultaneous Autoregressive Models,Sankhya, Vol 70-B, Part, Petrucci A, Salvati N (006) Small area estimation for spatial correlation in watershed erosion assessment. J Agric Biol Environ Stat (), 69 8 Pratesi M, Salvati N (004) Spatial EBLUP in agricultural survey. An application based on census data. Report no 56, Department of Statistics and Mathematics, University of Pisa Pratesi M, Salvati N (005) Small area estimation: the EBLUP estimator with autoregressive random area effects. Report no 6, Department of Statistics and Mathematics, University of Pisa

36 Literature Pratesi M., Salvati N. (008), Small area estimation: the EBLUP estimator based on spatially correlated random area effects. Statistical Methods and Applications, 7, No, 3-4 Rao, J.N.K. (003). Small Area Estimation, Wiley Interscience, Hoboken, New Jersey. Saei A, Chambers R (005) Small area estimation under linear and generalized linear mixed models with time and area effects. Southampton Statistical Sciences Research Institute, WP M03/5, Southampton SAMPLE project (00) Deliverables and 6, Final small area estimation developments and simulation results, SAMPLE project (0) Deliverable. Software on small area estimation, 9-36 Singh B.B., Shukla K., Kundu D. (005) Spatial-temporal models in small area estimation. Survey Methodology 3(), Spiegelhalter, D.J., Thomas, A., Best, N., And Lunn, D. (003). WinBUGS User Manual, Version.4. Sturtz, S., Ligges, U., And Gelman, A. (005). RWinBUGS: A Package for Running WinBUGS from R., Journal of Statistical Software, (3), -6. Vogt, M. (00), Bayesian Spatial Modeling: Propriety and Applications to Small Area Estimation with Focus on the German Census 0, PhD Thesis, University of Trier

SMALL AREA ESTIMATION OF INCOME UNDER SPATIAL SAR MODEL

SMALL AREA ESTIMATION OF INCOME UNDER SPATIAL SAR MODEL STATISTICS IN TRANSITION new series, September 2016 365 STATISTICS IN TRANSITION new series, September 2016 Vol. 17, No. 3, pp. 365 390 SMALL AREA ESTIMATION OF INCOME UNDER SPATIAL SAR MODEL Jan Kubacki

More information

ESTIMATION OF SMALL AREA CHARACTERISTICS USING MULTIVARIATE RAO-YU MODEL

ESTIMATION OF SMALL AREA CHARACTERISTICS USING MULTIVARIATE RAO-YU MODEL STATISTICS IN TRANSITION new series, December 2017 725 STATISTICS IN TRANSITION new series, December 2017 Vol. 18, No. 4, pp. 725 742, DOI 10.21307/stattrans-2017-009 ESTIMATION OF SMALL AREA CHARACTERISTICS

More information

Optimum Spatial Weighted in Small Area Estimation

Optimum Spatial Weighted in Small Area Estimation Global Journal of Pure and Applied Mathematics. ISSN 0973-1768 Volume 12, Number 5 (2016), pp. 3977-3989 Research India Publications http://www.ripublication.com/gjpam.htm Optimum Spatial Weighted in Small

More information

Estimation of Mean Population in Small Area with Spatial Best Linear Unbiased Prediction Method

Estimation of Mean Population in Small Area with Spatial Best Linear Unbiased Prediction Method Journal of Physics: Conference Series PAPER OPEN ACCESS Estimation of Mean Population in Small Area with Spatial Best Linear Unbiased Prediction Method To cite this article: Syahril Ramadhan et al 2017

More information

Spatial Modeling and Prediction of County-Level Employment Growth Data

Spatial Modeling and Prediction of County-Level Employment Growth Data Spatial Modeling and Prediction of County-Level Employment Growth Data N. Ganesh Abstract For correlated sample survey estimates, a linear model with covariance matrix in which small areas are grouped

More information

Small Area Estimation via Multivariate Fay-Herriot Models with Latent Spatial Dependence

Small Area Estimation via Multivariate Fay-Herriot Models with Latent Spatial Dependence Small Area Estimation via Multivariate Fay-Herriot Models with Latent Spatial Dependence Aaron T. Porter 1, Christopher K. Wikle 2, Scott H. Holan 2 arxiv:1310.7211v1 [stat.me] 27 Oct 2013 Abstract The

More information

Small Area Estimates of Poverty Incidence in the State of Uttar Pradesh in India

Small Area Estimates of Poverty Incidence in the State of Uttar Pradesh in India Small Area Estimates of Poverty Incidence in the State of Uttar Pradesh in India Hukum Chandra Indian Agricultural Statistics Research Institute, New Delhi Email: hchandra@iasri.res.in Acknowledgments

More information

Bayesian Networks in Educational Assessment

Bayesian Networks in Educational Assessment Bayesian Networks in Educational Assessment Estimating Parameters with MCMC Bayesian Inference: Expanding Our Context Roy Levy Arizona State University Roy.Levy@asu.edu 2017 Roy Levy MCMC 1 MCMC 2 Posterior

More information

Package sae. R topics documented: July 8, 2015

Package sae. R topics documented: July 8, 2015 Type Package Title Small Area Estimation Version 1.1 Date 2015-08-07 Author Isabel Molina, Yolanda Marhuenda Package sae July 8, 2015 Maintainer Yolanda Marhuenda Depends stats, nlme,

More information

Analysing Spatial Data in R Worked examples: Small Area Estimation

Analysing Spatial Data in R Worked examples: Small Area Estimation Analysing Spatial Data in R Worked examples: Small Area Estimation Virgilio Gómez-Rubio Department of Epidemiology and Public Heath Imperial College London London, UK 31 August 2007 Small Area Estimation

More information

ESTP course on Small Area Estimation

ESTP course on Small Area Estimation ESTP course on Small Area Estimation Statistics Finland, Helsinki, 29 September 2 October 2014 Topic 1: Introduction to small area estimation Risto Lehtonen, University of Helsinki Lecture topics: Monday

More information

Small Area Estimation with Uncertain Random Effects

Small Area Estimation with Uncertain Random Effects Small Area Estimation with Uncertain Random Effects Gauri Sankar Datta 1 2 3 and Abhyuday Mandal 4 Abstract Random effects models play an important role in model-based small area estimation. Random effects

More information

Model-based Estimation of Poverty Indicators for Small Areas: Overview. J. N. K. Rao Carleton University, Ottawa, Canada

Model-based Estimation of Poverty Indicators for Small Areas: Overview. J. N. K. Rao Carleton University, Ottawa, Canada Model-based Estimation of Poverty Indicators for Small Areas: Overview J. N. K. Rao Carleton University, Ottawa, Canada Isabel Molina Universidad Carlos III de Madrid, Spain Paper presented at The First

More information

Small Area Modeling of County Estimates for Corn and Soybean Yields in the US

Small Area Modeling of County Estimates for Corn and Soybean Yields in the US Small Area Modeling of County Estimates for Corn and Soybean Yields in the US Matt Williams National Agricultural Statistics Service United States Department of Agriculture Matt.Williams@nass.usda.gov

More information

Spatial Analysis of Incidence Rates: A Bayesian Approach

Spatial Analysis of Incidence Rates: A Bayesian Approach Spatial Analysis of Incidence Rates: A Bayesian Approach Silvio A. da Silva, Luiz L.M. Melo and Ricardo Ehlers July 2004 Abstract Spatial models have been used in many fields of science where the data

More information

Fully Bayesian Spatial Analysis of Homicide Rates.

Fully Bayesian Spatial Analysis of Homicide Rates. Fully Bayesian Spatial Analysis of Homicide Rates. Silvio A. da Silva, Luiz L.M. Melo and Ricardo S. Ehlers Universidade Federal do Paraná, Brazil Abstract Spatial models have been used in many fields

More information

Spatial M-quantile Models for Small Area Estimation

Spatial M-quantile Models for Small Area Estimation University of Wollongong Research Online Centre for Statistical & Survey Methodology Working Paper Series Faculty of Engineering and Information Sciences 2008 Spatial M-quantile Models for Small Area Estimation

More information

WinBUGS : part 2. Bruno Boulanger Jonathan Jaeger Astrid Jullion Philippe Lambert. Gabriele, living with rheumatoid arthritis

WinBUGS : part 2. Bruno Boulanger Jonathan Jaeger Astrid Jullion Philippe Lambert. Gabriele, living with rheumatoid arthritis WinBUGS : part 2 Bruno Boulanger Jonathan Jaeger Astrid Jullion Philippe Lambert Gabriele, living with rheumatoid arthritis Agenda 2! Hierarchical model: linear regression example! R2WinBUGS Linear Regression

More information

Bayesian Estimation of Expected Cell Counts by Using R

Bayesian Estimation of Expected Cell Counts by Using R Bayesian Estimation of Expected Cell Counts by Using R Haydar Demirhan 1 and Canan Hamurkaroglu 2 Department of Statistics, Hacettepe University, Beytepe, 06800, Ankara, Turkey Abstract In this article,

More information

The Bayesian Approach to Multi-equation Econometric Model Estimation

The Bayesian Approach to Multi-equation Econometric Model Estimation Journal of Statistical and Econometric Methods, vol.3, no.1, 2014, 85-96 ISSN: 2241-0384 (print), 2241-0376 (online) Scienpress Ltd, 2014 The Bayesian Approach to Multi-equation Econometric Model Estimation

More information

ESTP course on Small Area Estimation

ESTP course on Small Area Estimation Computational tools SAS and related software Kari Djerf Statistics Finland Contents Development of SAE software Built-in features in SAS Specific SAS macros 2 Development of SAE software Traditionally:

More information

An overview of Fay Herriot model with our package smallarea

An overview of Fay Herriot model with our package smallarea An overview of Fay Herriot model with our package smallarea Abhishek Nandy May 3, 2013 1 The Fay Herriot Model 1.1 Model Notations The Fay Herriot model can be written as follows: y i = x T i β + v i +

More information

Bayesian Prediction of Code Output. ASA Albuquerque Chapter Short Course October 2014

Bayesian Prediction of Code Output. ASA Albuquerque Chapter Short Course October 2014 Bayesian Prediction of Code Output ASA Albuquerque Chapter Short Course October 2014 Abstract This presentation summarizes Bayesian prediction methodology for the Gaussian process (GP) surrogate representation

More information

Saddlepoint-Based Bootstrap Inference in Dependent Data Settings

Saddlepoint-Based Bootstrap Inference in Dependent Data Settings Saddlepoint-Based Bootstrap Inference in Dependent Data Settings Alex Trindade Dept. of Mathematics & Statistics, Texas Tech University Rob Paige, Missouri University of Science and Technology Indika Wickramasinghe,

More information

A Comparative Study of Imputation Methods for Estimation of Missing Values of Per Capita Expenditure in Central Java

A Comparative Study of Imputation Methods for Estimation of Missing Values of Per Capita Expenditure in Central Java IOP Conference Series: Earth and Environmental Science PAPER OPEN ACCESS A Comparative Study of Imputation Methods for Estimation of Missing Values of Per Capita Expenditure in Central Java To cite this

More information

Metropolis-Hastings Algorithm

Metropolis-Hastings Algorithm Strength of the Gibbs sampler Metropolis-Hastings Algorithm Easy algorithm to think about. Exploits the factorization properties of the joint probability distribution. No difficult choices to be made to

More information

Advanced Statistical Modelling

Advanced Statistical Modelling Markov chain Monte Carlo (MCMC) Methods and Their Applications in Bayesian Statistics School of Technology and Business Studies/Statistics Dalarna University Borlänge, Sweden. Feb. 05, 2014. Outlines 1

More information

Selection of small area estimation method for Poverty Mapping: A Conceptual Framework

Selection of small area estimation method for Poverty Mapping: A Conceptual Framework Selection of small area estimation method for Poverty Mapping: A Conceptual Framework Sumonkanti Das National Institute for Applied Statistics Research Australia University of Wollongong The First Asian

More information

PIER HLM Course July 30, 2011 Howard Seltman. Discussion Guide for Bayes and BUGS

PIER HLM Course July 30, 2011 Howard Seltman. Discussion Guide for Bayes and BUGS PIER HLM Course July 30, 2011 Howard Seltman Discussion Guide for Bayes and BUGS 1. Classical Statistics is based on parameters as fixed unknown values. a. The standard approach is to try to discover,

More information

On Modifications to Linking Variance Estimators in the Fay-Herriot Model that Induce Robustness

On Modifications to Linking Variance Estimators in the Fay-Herriot Model that Induce Robustness Statistics and Applications {ISSN 2452-7395 (online)} Volume 16 No. 1, 2018 (New Series), pp 289-303 On Modifications to Linking Variance Estimators in the Fay-Herriot Model that Induce Robustness Snigdhansu

More information

Bayesian Methods for Machine Learning

Bayesian Methods for Machine Learning Bayesian Methods for Machine Learning CS 584: Big Data Analytics Material adapted from Radford Neal s tutorial (http://ftp.cs.utoronto.ca/pub/radford/bayes-tut.pdf), Zoubin Ghahramni (http://hunch.net/~coms-4771/zoubin_ghahramani_bayesian_learning.pdf),

More information

Contextual Effects in Modeling for Small Domains

Contextual Effects in Modeling for Small Domains University of Wollongong Research Online Applied Statistics Education and Research Collaboration (ASEARC) - Conference Papers Faculty of Engineering and Information Sciences 2011 Contextual Effects in

More information

Estimation of Complex Small Area Parameters with Application to Poverty Indicators

Estimation of Complex Small Area Parameters with Application to Poverty Indicators 1 Estimation of Complex Small Area Parameters with Application to Poverty Indicators J.N.K. Rao School of Mathematics and Statistics, Carleton University (Joint work with Isabel Molina from Universidad

More information

R Demonstration ANCOVA

R Demonstration ANCOVA R Demonstration ANCOVA Objective: The purpose of this week s session is to demonstrate how to perform an analysis of covariance (ANCOVA) in R, and how to plot the regression lines for each level of the

More information

eqr094: Hierarchical MCMC for Bayesian System Reliability

eqr094: Hierarchical MCMC for Bayesian System Reliability eqr094: Hierarchical MCMC for Bayesian System Reliability Alyson G. Wilson Statistical Sciences Group, Los Alamos National Laboratory P.O. Box 1663, MS F600 Los Alamos, NM 87545 USA Phone: 505-667-9167

More information

Markov Chain Monte Carlo in Practice

Markov Chain Monte Carlo in Practice Markov Chain Monte Carlo in Practice Edited by W.R. Gilks Medical Research Council Biostatistics Unit Cambridge UK S. Richardson French National Institute for Health and Medical Research Vilejuif France

More information

Bayesian Inference. Chapter 4: Regression and Hierarchical Models

Bayesian Inference. Chapter 4: Regression and Hierarchical Models Bayesian Inference Chapter 4: Regression and Hierarchical Models Conchi Ausín and Mike Wiper Department of Statistics Universidad Carlos III de Madrid Advanced Statistics and Data Mining Summer School

More information

Bootstrap Approach to Comparison of Alternative Methods of Parameter Estimation of a Simultaneous Equation Model

Bootstrap Approach to Comparison of Alternative Methods of Parameter Estimation of a Simultaneous Equation Model Bootstrap Approach to Comparison of Alternative Methods of Parameter Estimation of a Simultaneous Equation Model Olubusoye, O. E., J. O. Olaomi, and O. O. Odetunde Abstract A bootstrap simulation approach

More information

Multivariate area level models for small area estimation. a

Multivariate area level models for small area estimation. a Multivariate area level models for small area estimation. a a In collaboration with Roberto Benavent Domingo Morales González d.morales@umh.es Universidad Miguel Hernández de Elche Multivariate area level

More information

Supplementary Appendix to Dynamic Asset Price Jumps: the Performance of High Frequency Tests and Measures, and the Robustness of Inference

Supplementary Appendix to Dynamic Asset Price Jumps: the Performance of High Frequency Tests and Measures, and the Robustness of Inference Supplementary Appendix to Dynamic Asset Price Jumps: the Performance of High Frequency Tests and Measures, and the Robustness of Inference Worapree Maneesoonthorn, Gael M. Martin, Catherine S. Forbes August

More information

A Comparison of model-based methods for Small Area Estimation

A Comparison of model-based methods for Small Area Estimation A Comparison of model-based methods for Small Area Estimation N. Best 1, S. Richardson 1, P. Clarke 2 and V. Gómez-Rubio 1 1 Department of Epidemiology and Public Health, Imperial College London St. Mary

More information

Example: Ground Motion Attenuation

Example: Ground Motion Attenuation Example: Ground Motion Attenuation Problem: Predict the probability distribution for Peak Ground Acceleration (PGA), the level of ground shaking caused by an earthquake Earthquake records are used to update

More information

DIC: Deviance Information Criterion

DIC: Deviance Information Criterion (((( Welcome Page Latest News DIC: Deviance Information Criterion Contact us/bugs list WinBUGS New WinBUGS examples FAQs DIC GeoBUGS DIC (Deviance Information Criterion) is a Bayesian method for model

More information

Small Area Estimation for Survey Data Analysis Using SAS Software

Small Area Estimation for Survey Data Analysis Using SAS Software Small Area Estimation for Survey Data Analysis Using SAS Software Pushpal K Mukhopadhyay and Allen McDowell, SAS Institute Inc., Cary, NC ABSTRACT Small area estimation is important in survey analysis

More information

Small Area Confidence Bounds on Small Cell Proportions in Survey Populations

Small Area Confidence Bounds on Small Cell Proportions in Survey Populations Small Area Confidence Bounds on Small Cell Proportions in Survey Populations Aaron Gilary, Jerry Maples, U.S. Census Bureau U.S. Census Bureau Eric V. Slud, U.S. Census Bureau Univ. Maryland College Park

More information

Small Area Estimation in R with Application to Mexican Income Data

Small Area Estimation in R with Application to Mexican Income Data with Application to Mexican Income Data Ann-Kristin Kreutzmann 1, Sören Pannier 1 Natalia Rojas-Perilla 1, Timo Schmid 1 Matthias Templ 2 Nikos Tzavidis 3 1 Freie Universität Berlin 2 Zürcher Hochschule

More information

Analyzing Spatial Panel Data of Cigarette Demand: A Bayesian Hierarchical Modeling Approach

Analyzing Spatial Panel Data of Cigarette Demand: A Bayesian Hierarchical Modeling Approach Journal of Data Science 6(2008), 467-489 Analyzing Spatial Panel Data of Cigarette Demand: A Bayesian Hierarchical Modeling Approach Yanbing Zheng 1, Jun Zhu 2 and Dong Li 3 1 University of Kentucky, 2

More information

Robust Bayesian Regression

Robust Bayesian Regression Readings: Hoff Chapter 9, West JRSSB 1984, Fúquene, Pérez & Pericchi 2015 Duke University November 17, 2016 Body Fat Data: Intervals w/ All Data Response % Body Fat and Predictor Waist Circumference 95%

More information

Why Bayesian approaches? The average height of a rare plant

Why Bayesian approaches? The average height of a rare plant Why Bayesian approaches? The average height of a rare plant Estimation and comparison of averages is an important step in many ecological analyses and demographic models. In this demonstration you will

More information

Cross-sectional space-time modeling using ARNN(p, n) processes

Cross-sectional space-time modeling using ARNN(p, n) processes Cross-sectional space-time modeling using ARNN(p, n) processes W. Polasek K. Kakamu September, 006 Abstract We suggest a new class of cross-sectional space-time models based on local AR models and nearest

More information

Lecture 14 Bayesian Models for Spatio-Temporal Data

Lecture 14 Bayesian Models for Spatio-Temporal Data Lecture 14 Bayesian Models for Spatio-Temporal Data Dennis Sun Stats 253 August 13, 2014 Outline of Lecture 1 Recap of Bayesian Models 2 Empirical Bayes 3 Case 1: Long-Lead Forecasting of Sea Surface Temperatures

More information

Part 8: GLMs and Hierarchical LMs and GLMs

Part 8: GLMs and Hierarchical LMs and GLMs Part 8: GLMs and Hierarchical LMs and GLMs 1 Example: Song sparrow reproductive success Arcese et al., (1992) provide data on a sample from a population of 52 female song sparrows studied over the course

More information

Approaches for Multiple Disease Mapping: MCAR and SANOVA

Approaches for Multiple Disease Mapping: MCAR and SANOVA Approaches for Multiple Disease Mapping: MCAR and SANOVA Dipankar Bandyopadhyay Division of Biostatistics, University of Minnesota SPH April 22, 2015 1 Adapted from Sudipto Banerjee s notes SANOVA vs MCAR

More information

Plausible Values for Latent Variables Using Mplus

Plausible Values for Latent Variables Using Mplus Plausible Values for Latent Variables Using Mplus Tihomir Asparouhov and Bengt Muthén August 21, 2010 1 1 Introduction Plausible values are imputed values for latent variables. All latent variables can

More information

Bayesian modelling of real GDP rate in Romania

Bayesian modelling of real GDP rate in Romania Bayesian modelling of real GDP rate in Romania Mihaela SIMIONESCU 1* 1 Romanian Academy, Institute for Economic Forecasting, Bucharest, Calea 13 Septembrie, no. 13, District 5, Bucharest 1 Abstract The

More information

Default Priors and Effcient Posterior Computation in Bayesian

Default Priors and Effcient Posterior Computation in Bayesian Default Priors and Effcient Posterior Computation in Bayesian Factor Analysis January 16, 2010 Presented by Eric Wang, Duke University Background and Motivation A Brief Review of Parameter Expansion Literature

More information

Measurement error as missing data: the case of epidemiologic assays. Roderick J. Little

Measurement error as missing data: the case of epidemiologic assays. Roderick J. Little Measurement error as missing data: the case of epidemiologic assays Roderick J. Little Outline Discuss two related calibration topics where classical methods are deficient (A) Limit of quantification methods

More information

ARIC Manuscript Proposal # PC Reviewed: _9/_25_/06 Status: A Priority: _2 SC Reviewed: _9/_25_/06 Status: A Priority: _2

ARIC Manuscript Proposal # PC Reviewed: _9/_25_/06 Status: A Priority: _2 SC Reviewed: _9/_25_/06 Status: A Priority: _2 ARIC Manuscript Proposal # 1186 PC Reviewed: _9/_25_/06 Status: A Priority: _2 SC Reviewed: _9/_25_/06 Status: A Priority: _2 1.a. Full Title: Comparing Methods of Incorporating Spatial Correlation in

More information

Lognormal Measurement Error in Air Pollution Health Effect Studies

Lognormal Measurement Error in Air Pollution Health Effect Studies Lognormal Measurement Error in Air Pollution Health Effect Studies Richard L. Smith Department of Statistics and Operations Research University of North Carolina, Chapel Hill rls@email.unc.edu Presentation

More information

Bayesian Correlated Factor Analysis for Spatial Data

Bayesian Correlated Factor Analysis for Spatial Data Bayesian Correlated Factor Analysis for Spatial Data Maura Mezzetti Dipartimento di Studi Economico-Finanziari e Metodi Quantitativi, Universita Tor Vergata, Roma e-mail: maura.mezzetti@uniroma2.it Abstract:

More information

Nonparametric Small Area Estimation via M-quantile Regression using Penalized Splines

Nonparametric Small Area Estimation via M-quantile Regression using Penalized Splines Nonparametric Small Estimation via M-quantile Regression using Penalized Splines Monica Pratesi 10 August 2008 Abstract The demand of reliable statistics for small areas, when only reduced sizes of the

More information

A Bayesian Approach for Sample Size Determination in Method Comparison Studies

A Bayesian Approach for Sample Size Determination in Method Comparison Studies A Bayesian Approach for Sample Size Determination in Method Comparison Studies Kunshan Yin a, Pankaj K. Choudhary a,1, Diana Varghese b and Steven R. Goodman b a Department of Mathematical Sciences b Department

More information

Lecture 5: Spatial probit models. James P. LeSage University of Toledo Department of Economics Toledo, OH

Lecture 5: Spatial probit models. James P. LeSage University of Toledo Department of Economics Toledo, OH Lecture 5: Spatial probit models James P. LeSage University of Toledo Department of Economics Toledo, OH 43606 jlesage@spatial-econometrics.com March 2004 1 A Bayesian spatial probit model with individual

More information

Markov Chain Monte Carlo methods

Markov Chain Monte Carlo methods Markov Chain Monte Carlo methods By Oleg Makhnin 1 Introduction a b c M = d e f g h i 0 f(x)dx 1.1 Motivation 1.1.1 Just here Supresses numbering 1.1.2 After this 1.2 Literature 2 Method 2.1 New math As

More information

Time-series small area estimation for unemployment based on a rotating panel survey

Time-series small area estimation for unemployment based on a rotating panel survey Discussion Paper Time-series small area estimation for unemployment based on a rotating panel survey The views expressed in this paper are those of the author and do not necessarily relect the policies

More information

Multivariate Normal & Wishart

Multivariate Normal & Wishart Multivariate Normal & Wishart Hoff Chapter 7 October 21, 2010 Reading Comprehesion Example Twenty-two children are given a reading comprehsion test before and after receiving a particular instruction method.

More information

This module is part of the. Memobust Handbook. on Methodology of Modern Business Statistics

This module is part of the. Memobust Handbook. on Methodology of Modern Business Statistics This moule is part of the Memobust Hanbook on Methoology of Moern Business Statistics 26 March 2014 Metho: EBLUP Unit Level for Small Area Estimation Contents General section... 3 1. Summary... 3 2. General

More information

Small Domains Estimation and Poverty Indicators. Carleton University, Ottawa, Canada

Small Domains Estimation and Poverty Indicators. Carleton University, Ottawa, Canada Small Domains Estimation and Poverty Indicators J. N. K. Rao Carleton University, Ottawa, Canada Invited paper for presentation at the International Seminar Population Estimates and Projections: Methodologies,

More information

Bayesian Inference. Chapter 4: Regression and Hierarchical Models

Bayesian Inference. Chapter 4: Regression and Hierarchical Models Bayesian Inference Chapter 4: Regression and Hierarchical Models Conchi Ausín and Mike Wiper Department of Statistics Universidad Carlos III de Madrid Master in Business Administration and Quantitative

More information

A Lightweight Guide on Gibbs Sampling and JAGS

A Lightweight Guide on Gibbs Sampling and JAGS A Lightweight Guide on Gibbs Sampling and JAGS Gianpaolo Coro Istituto di Scienza e Tecnologie dell Informazione A. Faedo, Pisa, Italy 1 Summary In this document we give some insight about how Gibbs Sampling

More information

Estimating Income Distributions Using a Mixture of Gamma. Densities

Estimating Income Distributions Using a Mixture of Gamma. Densities Estimating Income Distributions Using a Mixture of Gamma Densities Duangkamon Chotikapanich Monash University William E Griffiths University of Melbourne 3 May 2007 Abstract The estimation of income distributions

More information

Stat 544 Final Exam. May 2, I have neither given nor received unauthorized assistance on this examination.

Stat 544 Final Exam. May 2, I have neither given nor received unauthorized assistance on this examination. Stat 544 Final Exam May, 006 I have neither given nor received unauthorized assistance on this examination. signature date 1 1. Below is a directed acyclic graph that represents a joint distribution for

More information

LFS quarterly small area estimation of youth unemployment at provincial level

LFS quarterly small area estimation of youth unemployment at provincial level LFS quarterly small area estimation of youth unemployment at provincial level Stima trimestrale della disoccupazione giovanile a livello provinciale su dati RFL Michele D Aló, Stefano Falorsi, Silvia Loriga

More information

Models for spatial data (cont d) Types of spatial data. Types of spatial data (cont d) Hierarchical models for spatial data

Models for spatial data (cont d) Types of spatial data. Types of spatial data (cont d) Hierarchical models for spatial data Hierarchical models for spatial data Based on the book by Banerjee, Carlin and Gelfand Hierarchical Modeling and Analysis for Spatial Data, 2004. We focus on Chapters 1, 2 and 5. Geo-referenced data arise

More information

Making rating curves - the Bayesian approach

Making rating curves - the Bayesian approach Making rating curves - the Bayesian approach Rating curves what is wanted? A best estimate of the relationship between stage and discharge at a given place in a river. The relationship should be on the

More information

NONLINEAR APPLICATIONS OF MARKOV CHAIN MONTE CARLO

NONLINEAR APPLICATIONS OF MARKOV CHAIN MONTE CARLO NONLINEAR APPLICATIONS OF MARKOV CHAIN MONTE CARLO by Gregois Lee, B.Sc.(ANU), B.Sc.Hons(UTas) Submitted in fulfilment of the requirements for the Degree of Doctor of Philosophy Department of Mathematics

More information

Bayesian linear regression

Bayesian linear regression Bayesian linear regression Linear regression is the basis of most statistical modeling. The model is Y i = X T i β + ε i, where Y i is the continuous response X i = (X i1,..., X ip ) T is the corresponding

More information

Small area solutions for the analysis of pollution data TIES Daniela Cocchi, Enrico Fabrizi, Carlo Trivisano. Università di Bologna

Small area solutions for the analysis of pollution data TIES Daniela Cocchi, Enrico Fabrizi, Carlo Trivisano. Università di Bologna Small area solutions for the analysis of pollution data Daniela Cocchi, Enrico Fabrizi, Carlo Trivisano Università di Bologna TIES 00 Genova, June nd 00 Summary The small area problem in the environmental

More information

Bayesian Econometrics - Computer section

Bayesian Econometrics - Computer section Bayesian Econometrics - Computer section Leandro Magnusson Department of Economics Brown University Leandro Magnusson@brown.edu http://www.econ.brown.edu/students/leandro Magnusson/ April 26, 2006 Preliminary

More information

Slice Sampling with Adaptive Multivariate Steps: The Shrinking-Rank Method

Slice Sampling with Adaptive Multivariate Steps: The Shrinking-Rank Method Slice Sampling with Adaptive Multivariate Steps: The Shrinking-Rank Method Madeleine B. Thompson Radford M. Neal Abstract The shrinking rank method is a variation of slice sampling that is efficient at

More information

Statistics in Environmental Research (BUC Workshop Series) II Problem sheet - WinBUGS - SOLUTIONS

Statistics in Environmental Research (BUC Workshop Series) II Problem sheet - WinBUGS - SOLUTIONS Statistics in Environmental Research (BUC Workshop Series) II Problem sheet - WinBUGS - SOLUTIONS 1. (a) The posterior mean estimate of α is 14.27, and the posterior mean for the standard deviation of

More information

U.P., India Dr.Vijay Kumar, Govt. College for Women, M.A. Road Srinagar (J & K), India ABSTRACT

U.P., India Dr.Vijay Kumar, Govt. College for Women, M.A. Road Srinagar (J & K), India ABSTRACT BAYESIAN ESTIMATION OF THE PARAMETER OF GENERALIZED EXPONENTIAL DISTRIBUTION USING MARKOV CHAIN MONTE CARLO METHOD IN OPEN BUGS FOR INFORMATIVE SET OF PRIORS Dr. Mahmood Alam Khan, Dr. Aijaz Ahmed Hakkak,

More information

The lmm Package. May 9, Description Some improved procedures for linear mixed models

The lmm Package. May 9, Description Some improved procedures for linear mixed models The lmm Package May 9, 2005 Version 0.3-4 Date 2005-5-9 Title Linear mixed models Author Original by Joseph L. Schafer . Maintainer Jing hua Zhao Description Some improved

More information

Advanced Machine Learning

Advanced Machine Learning Advanced Machine Learning Nonparametric Bayesian Models --Learning/Reasoning in Open Possible Worlds Eric Xing Lecture 7, August 4, 2009 Reading: Eric Xing Eric Xing @ CMU, 2006-2009 Clustering Eric Xing

More information

Online Appendix to: Crises and Recoveries in an Empirical Model of. Consumption Disasters

Online Appendix to: Crises and Recoveries in an Empirical Model of. Consumption Disasters Online Appendix to: Crises and Recoveries in an Empirical Model of Consumption Disasters Emi Nakamura Columbia University Robert Barro Harvard University Jón Steinsson Columbia University José Ursúa Harvard

More information

Package zic. August 22, 2017

Package zic. August 22, 2017 Package zic August 22, 2017 Version 0.9.1 Date 2017-08-22 Title Bayesian Inference for Zero-Inflated Count Models Author Markus Jochmann Maintainer Markus Jochmann

More information

Spatial Fay-Herriot Models for Small Area Estimation with Functional Covariates

Spatial Fay-Herriot Models for Small Area Estimation with Functional Covariates Spatial Fay-Herriot Models for Small Area Estimation with Functional Covariates Aaron T. Porter 1, Scott H. Holan 2, Christopher K. Wikle 2, Noel Cressie 2, 3 arxiv:1303.6668v3 [stat.me] 9 May 2014 Abstract

More information

CTDL-Positive Stable Frailty Model

CTDL-Positive Stable Frailty Model CTDL-Positive Stable Frailty Model M. Blagojevic 1, G. MacKenzie 2 1 Department of Mathematics, Keele University, Staffordshire ST5 5BG,UK and 2 Centre of Biostatistics, University of Limerick, Ireland

More information

MCMC algorithms for fitting Bayesian models

MCMC algorithms for fitting Bayesian models MCMC algorithms for fitting Bayesian models p. 1/1 MCMC algorithms for fitting Bayesian models Sudipto Banerjee sudiptob@biostat.umn.edu University of Minnesota MCMC algorithms for fitting Bayesian models

More information

STAT 425: Introduction to Bayesian Analysis

STAT 425: Introduction to Bayesian Analysis STAT 425: Introduction to Bayesian Analysis Marina Vannucci Rice University, USA Fall 2017 Marina Vannucci (Rice University, USA) Bayesian Analysis (Part 2) Fall 2017 1 / 19 Part 2: Markov chain Monte

More information

Kazuhiko Kakamu Department of Economics Finance, Institute for Advanced Studies. Abstract

Kazuhiko Kakamu Department of Economics Finance, Institute for Advanced Studies. Abstract Bayesian Estimation of A Distance Functional Weight Matrix Model Kazuhiko Kakamu Department of Economics Finance, Institute for Advanced Studies Abstract This paper considers the distance functional weight

More information

Multivariate beta regression with application to small area estimation

Multivariate beta regression with application to small area estimation Multivariate beta regression with application to small area estimation Debora Ferreira de Souza debora@dme.ufrj.br Fernando Antônio da Silva Moura fmoura@im.ufrj.br Departamento de Métodos Estatísticos

More information

Introduction to Survey Data Integration

Introduction to Survey Data Integration Introduction to Survey Data Integration Jae-Kwang Kim Iowa State University May 20, 2014 Outline 1 Introduction 2 Survey Integration Examples 3 Basic Theory for Survey Integration 4 NASS application 5

More information

Spatial Econometrics

Spatial Econometrics Spatial Econometrics Lecture 5: Single-source model of spatial regression. Combining GIS and regional analysis (5) Spatial Econometrics 1 / 47 Outline 1 Linear model vs SAR/SLM (Spatial Lag) Linear model

More information

Monte Carlo in Bayesian Statistics

Monte Carlo in Bayesian Statistics Monte Carlo in Bayesian Statistics Matthew Thomas SAMBa - University of Bath m.l.thomas@bath.ac.uk December 4, 2014 Matthew Thomas (SAMBa) Monte Carlo in Bayesian Statistics December 4, 2014 1 / 16 Overview

More information

Bayesian modelling. Hans-Peter Helfrich. University of Bonn. Theodor-Brinkmann-Graduate School

Bayesian modelling. Hans-Peter Helfrich. University of Bonn. Theodor-Brinkmann-Graduate School Bayesian modelling Hans-Peter Helfrich University of Bonn Theodor-Brinkmann-Graduate School H.-P. Helfrich (University of Bonn) Bayesian modelling Brinkmann School 1 / 22 Overview 1 Bayesian modelling

More information

Areal data models. Spatial smoothers. Brook s Lemma and Gibbs distribution. CAR models Gaussian case Non-Gaussian case

Areal data models. Spatial smoothers. Brook s Lemma and Gibbs distribution. CAR models Gaussian case Non-Gaussian case Areal data models Spatial smoothers Brook s Lemma and Gibbs distribution CAR models Gaussian case Non-Gaussian case SAR models Gaussian case Non-Gaussian case CAR vs. SAR STAR models Inference for areal

More information

Information. Hierarchical Models - Statistical Methods. References. Outline

Information. Hierarchical Models - Statistical Methods. References. Outline Information Hierarchical Models - Statistical Methods Sarah Filippi 1 University of Oxford Hilary Term 2015 Webpage: http://www.stats.ox.ac.uk/~filippi/msc_ hierarchicalmodels_2015.html Lectures: Week

More information

Non-parametric bootstrap mean squared error estimation for M-quantile estimates of small area means, quantiles and poverty indicators

Non-parametric bootstrap mean squared error estimation for M-quantile estimates of small area means, quantiles and poverty indicators Non-parametric bootstrap mean squared error estimation for M-quantile estimates of small area means, quantiles and poverty indicators Stefano Marchetti 1 Nikos Tzavidis 2 Monica Pratesi 3 1,3 Department

More information

Tutorial on Probabilistic Programming with PyMC3

Tutorial on Probabilistic Programming with PyMC3 185.A83 Machine Learning for Health Informatics 2017S, VU, 2.0 h, 3.0 ECTS Tutorial 02-04.04.2017 Tutorial on Probabilistic Programming with PyMC3 florian.endel@tuwien.ac.at http://hci-kdd.org/machine-learning-for-health-informatics-course

More information