A criterion-based PLS approach to SEM. Michel Tenenhaus (HEC Paris) Arthur Tenenhaus (SUPELEC)
|
|
- Thomasina Shelton
- 5 years ago
- Views:
Transcription
1 A criterion-based PLS approach to SEM Michel Tenenhaus (HEC Paris) Arthur Tenenhaus (SUPELEC)
2 Economic inequality and political instability Data from Russett (1964), in GIFI Economic inequality Agricultural inequality GINI : Inequality of land distributions FARM : % farmers that own half of the land (> 50) RENT : % farmers that rent all their land Industrial development GNPR : Gross national product per capita ($ 1955) LABO : % of labor force employed in agriculture Political instability INST : Instability of executive (45-61) ECKS : Nb of violent internal war incidents (46-61) DEAT : Nb of people killed as a result of civic group violence (50-62) D-STAB : Stable democracy D-UNST : Unstable democracy DICT : Dictatorship 2
3 Economic inequality and political instability (Data from Russett, 1964) Gini Farm Rent Gnpr Labo Inst Ecks Deat Demo Argentine Australie * Autriche France Yougoslavie = Stable democracy 2 = Unstable democracy 3 = Dictatorship 3
4 Agricultural inequality (X 1 ) A SEM model GINI INST FARM RENT ξ 1 C 12 = 0 C 13 = 1 ξ 3 ECKS DEAT D-STB GNPR LABO ξ 2 C 23 = 1 D-INS DICT Industrial development (X 2 ) Political instability (X 3 ) 4
5 Latent Variable outer estimation Y1 = X1w1 = w11gini + w12farm + w13rent Y2 = X 2w2 = w21gnpr + w22labo Y = X w = w INST + w ECKS + w DEATH w D-STB + w D-UNST w DICT 36 5
6 Some modified multi-block methods for SEM c jk = 1 if blocks are linked, 0 otherwise and c jj = 0 SUMCOR (Horst, 1961) SSQCOR (Mathes, 1993, Hanafi, 2004) SABSCOR (Mathes, 1993, Hanafi, 2004) MAXDIFF (Van de Geer, 1984) [SUMCOV] MAXDIFF B (Hanafi & Kiers, 2006) [SSQCOV] SABSCOV (Krämer, 2007) Max c Cor( X w, X w ) jk, jk, jk j j k k GENERALIZED CANONICAL CORRELATION ANALYSIS Max c Cor ( X w, X w ) jk, 2 jk j j k k Max c Cor( X w, X w ) All wj = 1 jk, jk j j k k Max c Cov( X w, X w ) GENERALIZED PLS REGRESSION All wj = 1 jk, jk j j k k Max c Cov ( X w, X w ) All wj = 1 jk, 2 jk j j k k Max c Cov( X w, X w ) jk j j k k 6
7 Covariance-based criteria for SEM c jk = 1 if blocks are linked, 0 otherwise and c jj = 0 SUMCOR-PLSPM SSQCOR-PLSPM SABSCOR-PLSPM SUMCOV-PLSPM SSQCOV-PLSPM SABSCOV-PLSPM Max c Cov( X w, X w ) All Var( X w ) = 1 j j jk, All Var( X w ) = 1 j j jk, jk j j k k Max c Cov ( X w, X w ) All Var( X w ) = 1 j j jk, 2 jk j j k k Max c Cov( X w, X w ) All = 1 jk, jk j j k k Max c Cov( X w, X w ) w j All = 1 jk, jk j j k k Max c Cov ( X w, X w ) w j All = 1 jk, 2 jk j j k k Max c Cov( X w, X w ) w j jk j j k k 7
8 A continuum approach j< k Maximize c g(cov( X w, X w )) subject to the constraints: 2 jk j j k k τ w + (1 τ ) Var( X w ) = 1, with 0 τ 1, i = 1,..., J i i i i i i where gx = x (Horst scheme) 2 ( ) x (Factorial scheme) x (Centroid scheme) 8
9 A general procedure to obtain critical points of the criteria Construct the Lagrangian function related to the optimization problem. Cancel the derivative of the Lagrangian function with respect to each w i. Use the Wold s procedure to solve the stationary equations ( Lohmöller s procedure). This procedure converges to a critical point of the criterion. The criterion increases at each step of the algorithm. 9
10 The general algorithm Y j = ± X j w j Initial step w j τ w Outer Estimation 2 + (1 τ ) Var( X w ) = 1 j j j j j Iterate until convergence Y 1 Y 2 Y J e j1 e j2 e jj Z j Inner estimation w j = 1 ' 1 ' [( τ ji+ (1 τ j) XX j j] XZ j j n 1 ZX[( τ I+ (1 τ ) XX] XZ n ' ' 1 ' j j j j j j j j Choice of weights e jh : - Horst : e jh = c jh - Centroid : e jh = c jh sign(cor(y h, Y j ) - Factorial : e jh = c jh Cov(Y h, Y j ) c 10 jh = 1 if blocks are linked, 0 otherwise and c jj = 0
11 Specific cases j< k Maximize c g(cov( X w, X w )) subject to the constraints: 2 jk j j k k τ w + (1 τ ) Var( X w ) = 1, with 0 τ 1, i = 1,..., J i i i i i Criterion SUMCOR SSQCOR SABSCOR Horst Factorial Centroid Scheme (g(x) = x) (g(x) = x 2 ) (g(x) = x ) Value of τ i i PLS Mode B With usual PLS-PM constraints: Var( X w ) = 1 i i 11
12 Specific cases j< k Maximize c g(cov( X w, X w )) subject to the constraints: 2 jk j j k k τ w + (1 τ ) Var( X w ) = 1, with 0 τ 1, i = 1,..., J i i i i i Criterion SUMCOV SSQCOV SABSCOV Horst Factorial Centroid Scheme (g(x) = x) (g(x) = x 2 ) (g(x) = x ) Value of τ i i PLS New Mode A With usual PLS regression constraint: w i = 1 12
13 I. PLS approach : 2 blocks X 1 ξ 1 ξ 2 X 2 Mode for weight calculation Y 1 = X 1 w 1 Y 2 = X 2 w 2 Method Deflation A A PLS regression of X 2 on X 1 On X 1 only (*) B A Redundancy analysis of X 2 with respect to X 1 On X 1 only A A Tucker Inter-Battery Factor Analysis On X 1 and X 2 B B Canonical correlation Analysis On X 1 and X 2 (*) Deflation: Working on residuals of the regression of X on the previous LV s in order to obtain orthogonal LV s. 13
14 PLS regression (2 components) Use of PLS-GRAPH dim 1 - Mode A for X - Mode A for Y - Deflate only X Max Cov( Xa, Yb) a = b = 1 = Max Cor( Xa, Yb)* Var( Xa)* Var( Yb) a = b = 1 dim 2 14
15 PLS Regression in SIMCA-P : PLS Scores australie 3 royaume-uni t[2] suisse canada états-unis belgique nouvelle zélande pays-bas suède luxembourg france West Germany norvège finlande danemark irlande argentine uruguay venezuela cuba espagne irak italie chili équateur pèrou autriche colombie grèce rép. dominicaine costa-rica salvador guatémala taiwan brésil panama honduras nicaragua égypte sud vietnam philippines inde libye bolivie -2 japon -3-4 pologne yougoslavie t[1] 15
16 Correlation loadings RENT 0.60 GINI FARM 0.40 GNPR DEMOSTB pc(corr)[comp. 2] INST DEMOINST DEAT ECKS DICTATURE LABO pc(corr)[comp. 1] 16
17 Redundancy analysis of X on Y (2 components) - Mode A for X - Mode B for Y - Deflate only X dim 1 Max a = Var ( Yb) = 1 = a = Var ( Yb) = 1 Cov( Xa, Yb) Max Cor( Xa, Yb)* Var( Xa) dim 2 2 Max Cor x j Yb Var ( Yb) = 1 j (, ) 17
18 Inter-battery factor analysis (2 components) - Mode A for X - Mode A for Y - Deflate both X and Y dim 1 Max Cov( Xa, Yb) a = b = 1 = Max Cor( Xa, Yb)* Var( Xa)* Var( Yb) a = b = 1 dim 2 18
19 Canonical correlation analysis (2 components) - Mode B for X - Mode B for Y - Deflate both X and Y dim 1 Max Var ( Xa) = Var ( Yb) = 1 Cov( Xa, Yb) dim 2 Max Var ( Xa) = Var ( Yb) = 1 Cor( Xa, Yb) 19
20 Barker & Rayens PLS DA - Mode A for X - Mode B for Y - Deflate only on X a Max = Var ( Yb) = 1 Cov( Xa, Yb) = a Max Cor( Xa, Yb)* Var( Xa) = Var ( Yb) = 1 Redundancy analysis of X with respect to Y 20
21 Barker & Rayens PLS DA Economic inequality vs Political regime Separation between political regimes is improved compared to PLS-DA. 21
22 II. Hierarchical model : J blocs X 1 ξ 1... ξ X 1.. X J X J ξ J Scheme for computation of the inner components Z j Computation of outer weights w j Horst Centroid Factorial Mode A SUMCOV SABSCOV SSQCOV Mode B SUMCOR SUMCOR SSQCOR (Carroll GCCA) Deflation : On original blocks and/or the super-block Generalized PLS regression Generalized CCA 22
23 III. Multi-block data analysis SABSCOR : PLS Mode B + Centroid scheme Use of XLSTAT-PLSPM Max Cor( X1w1, X 2w2) + Cor( X1w1, X 3w3) + Cor( X 2w2, X 3w3) = =
24 Multiblock data analysis SSQCOR : PLS Mode B + Factorial scheme Max Cor ( X1w1, X 2w2) + Cor ( X1w1, X 3w3) + Cor ( X 2w2, X 3w3) = + + =
25 Practice supports theory Mode B + Centroid Mode B + Factorial Agricultural inequality <--> Industrial development Agricultural inequality <--> Political Instability Industrial development <--> Political Instability SABSCOR * SSQCOR * * Criterion optimized by the method (checked on random initial weights) 25
26 IV. Structural Equation Modeling Agricultural inequality (X 1 ) GINI INST FARM RENT ξ 1 C 12 =0 C 13 =1 ξ 3 ECKS DEAT D-STB GNPR LABO ξ 2 C 23 =1 D-INS DICT Industrial development (X 2 ) C ij = 1 if ξ i and ξ j are connected = otherwise Political instability (X 3 ) 26
27 SABSCOR-PLSPM Mode B + Centroid scheme Y 1 = X 1 w 1 Y 3 = X 3 w 3 Y 2 = X 2 w 2 Max Cor( X1w1, X 3w3) + Cor( X 2w2, X 3w3) = =
28 SSQCOR-PLSPM Mode B + Factorial scheme Y 1 = X 1 w 1 Y 3 = X 3 w 3 Y 2 = X 2 w Max Cor ( X1w1, X 3w3) + Cor ( X 2w2, X 3w3) = =
29 Comparison between methods Mode B + Centroid scheme Mode B + Factorial scheme Cor( Y, Y ) + Cor( Y, Y ) * Cor ( Y, Y ) + Cor ( Y, Y ) * * Criterion optimized by the method (checked on random initial weights) Practice supports theory 29
30 Mode A + Centroid scheme The criterion optimized by the algorithm, if any, is unknown. 30
31 SABSCOV-PLSPM New Mode A + Centroid scheme weight Cor=.429 Cov= Cov=-1.69 Cor= Max Cov( X1w1, X 3w3) + Cov( X 2w2, X 3w3) = 2.69 One-step hierarchical PLS Regression 31
32 SABSCOV-PLSPM New Mode A + Factorial scheme weight Cor=.4276 Cov= Cov=-1.69 Cor= Max Cov ( X1w1, X 3w3) + Cov ( X 2w2, X 3w3) = 3.86 One-step hierarchical PLS Regression 32
33 Generalized Barker & Rayens PLS-DA SSQCOV-PLSPM New Mode A for X 1 and X 2 and Mode B for Y weight Cov Max Cor ( X1w1, X 3w3)* Var( X1w1) + Cor ( X 2w2, X 3w3)* Var( X 2w2) = 1.39 One-step hierarchical B&R PLS-DA
34 Generalized Barker & Rayens PLS-DA Industrial development Agricultural inequality
35 Conclusion In the PLS approach of Herman Wold, the constraint is: Var( X w ) = 1 j In the PLS regression of Svante Wold, the constraint is: This presentation unifies both approaches. j w j = 1 35
Structured data analysis with RGCCA. Arthur Tenenhaus 2015/02/13
Structured data analysis with RGCCA Arthur Tenenhaus 2015/02/13 Overview of the presentation Part I. Multi-block analysis Part II. Multi-block and Multi-way analysis p 1 p 2 p 3 p 1 p 2 p 3 n X 1 X...
More informationStructured data analysis
Structured data analysis with RGCCA Arthur Tenenhaus Séminaire de Statistique appliquée du CNAM, 16/03/2018 n p 1 p 2 p J X 1 X... 2 X J p (a) multiblock structure n 1 X 1 n 2 X 2... p 1 p 2 p J n X 1
More informationRole and treatment of categorical variables in PLS Path Models for Composite Indicators
Role and treatment of categorical variables in PLS Path Models for Composite Indicators Laura Trinchera 1,2 & Giorgio Russolillo 2! 1 Dipartimento di Studi sullo Sviluppo Economico, Università degli Studi
More informationRegularized Generalized Canonical Correlation Analysis Extended to Symbolic Data
Noname manuscript No. (will be inserted by the editor) Regularized Generalized Canonical Correlation Analysis Extended to Symbolic Data Received: date / Accepted: date Abstract Regularized Generalized
More informationGeneral structural model Part 1: Covariance structure and identification. Psychology 588: Covariance structure and factor models
General structural model Part 1: Covariance structure and identification Psychology 588: Covariance structure and factor models Latent variables 2 Interchangeably used: constructs --- substantively defined
More informationKernel Generalized Canonical Correlation Analysis
Kernel Generalized Canonical Correlation Analysis Arthur Tenenhaus To cite this version: Arthur Tenenhaus. Kernel Generalized Canonical Correlation Analysis. JdS 10, May 2010, Marseille, France. CD-ROM
More informationUnderstanding PLS path modeling parameters estimates: a study based on Monte Carlo simulation and customer satisfaction surveys
Understanding PLS path modeling parameters estimates: a study based on Monte Carlo simulation and customer satisfaction surveys Emmanuel Jakobowicz CEDRIC-CNAM - 292 rue Saint Martin - 75141 Paris Cedex
More informationChapter 2 PLS Path Modeling: From Foundations to Recent Developments and Open Issues for Model Assessment and Improvement
Chapter 2 PLS Path Modeling: From Foundations to Recent Developments and Open Issues for Model Assessment and Improvement Vincenzo Esposito Vinzi, Laura Trinchera, and Silvano Amato Abstract In this chapter
More informationAN INTRODUCTION TO PARTIAL LEAST SQUARES PATH MODELING
STA201 Analyse multivariée approfondie AN INTRODUCTION TO PARTIAL LEAST SQUARES PATH MODELING Giorgio Russolillo CNAM, Paris giorgio.russolillo@cnam.fr Component-Based vs Factor-Based Sructural Equation
More informationSupplementary Appendix for. Version: February 3, 2014
Supplementary Appendix for When Do Governments Resort to Election Violence? Version: February 3, 2014 This appendix provides supplementary information not included in the published draft. Supplementary
More informationPre-Processing of Dynamic Networks * Its Impact on Observed Communities and Other Analytics
Pre-Processing of Dynamic Networks * Its Impact on Observed Communities and Other nalytics Sofus. Macskassy Data Scientist, Facebook Unifying Theory and Experiment for Large-Scale Networks Simons Institute
More informationTopological Data Analysis for Brain Networks
Topological Data Analysis for Brain Networks Relating Functional Brain Network Topology to Clinical Measures of Behavior Bei Wang Phillips 1,2 University of Utah Joint work with Eleanor Wong 1,2, Sourabh
More informationInternational Student Enrollment Fall 2018 By CIP Code, Country of Citizenship, and Education Level Harpur College of Arts and Sciences
International Student Enrollment Fall 2018 By CIP Code, Country of Citizenship, and Education Level Harpur College of Arts and Sciences CIP Code Description Citizenship Graduate Undergrad Total 00.0000
More informationBootstrapping Self-Organizing Maps to Assess the Statistical Significance of Local Proximity
Bootstrapping Self-Organizing Maps to Assess the Statistical Significance of Local Proximity Eric de Bodt 1, Marie Cottrell 2 1 Université Catholique de Louvain, IAG-FIN, 1 pl. des Doyens, B-1348 Louvain-la-Neuve,
More informationDISTILLED SPIRITS - EXPORTS BY VALUE DECEMBER 2017
DISTILLED SPIRITS - EXPORTS BY VALUE DECEMBER 2017 U.S. COMMERCIAL EXPORTS OF DISTILLED SPIRITS - DECEMBER 2017 (U.S. DOLLARS) Da-Value-17-12 SUMMARY BY CLASS CLASS DECEMBER DECEMBER DOLLAR YTD YTD DOLLAR
More informationDynamic-Inner Partial Least Squares for Dynamic Data Modeling
Preprints of the 9th International Symposium on Advanced Control of Chemical Processes The International Federation of Automatic Control MoM.5 Dynamic-Inner Partial Least Squares for Dynamic Data Modeling
More informationPrinciple Components Analysis (PCA) Relationship Between a Linear Combination of Variables and Axes Rotation for PCA
Principle Components Analysis (PCA) Relationship Between a Linear Combination of Variables and Axes Rotation for PCA Principle Components Analysis: Uses one group of variables (we will call this X) In
More informationJosé Luis Palma, William Rose and Rüdiger Escobar Wolf Department of Geological Engineering and Sciences Michigan Technological University
Assessing the Volcanic Threat of Central American Volcanoes José Luis Palma, William Rose and Rüdiger Escobar Wolf Department of Geological Engineering and Sciences Michigan Technological University Outline
More informationMultilevel Component Analysis applied to the measurement of a complex product experience
Multilevel Component Analysis applied to the measurement of a complex product experience Boucon, C.A., Petit-Jublot, C.E.F., Groeneschild C., Dijksterhuis, G.B. Outline Background Introduction to Simultaneous
More informationJavier Corrales Professor of Political Science Amherst College, Amherst, MA September 2013
Javier Corrales Professor of Political Science Amherst College, Amherst, MA jcorrales@amherst.edu September 2013 Sections I. Progress thus far II. Homophobia across and within III. Conventional independent
More informationStructural Equation Modeling and simultaneous clustering through the Partial Least Squares algorithm
Structural Equation Modeling and simultaneous clustering through the Partial Least Squares algorithm Mario Fordellone a, Maurizio Vichi a a Sapienza, University of Rome Abstract The identification of different
More informationUSDA Dairy Import License Circular for 2018
USDA Dairy Import License Circular for 2018 Commodity/Note Country Name TRQ Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Grand Total Non-Cheese 21,864,781 624,064 651,121 432,669 901,074 1,202,567 907,493
More informationDoes Corruption Persist In Sub-Saharan Africa?
Int Adv Econ Res (2009) 15:336 350 DOI 10.1007/s11294-009-9210-2 ORIGINAL PAPER Does Corruption Persist In Sub-Saharan Africa? Nicole Bissessar Published online: 12 June 2009 # International Atlantic Economic
More informationSolow model: Convergence
Solow model: Convergence Per capita income k(0)>k* Assume same s, δ, & n, but no technical progress y* k(0)=k* k(0) k Assume same s, δ, &
More informationChapter 8 - Appendixes
Chapter 8 - Appendixes Appendix 8.. Individual Preferences for Growth, Environment, and Income Distribution Funds to be invested in projects that Funds to be invested in projects to Funds to be invested
More informationarxiv: v1 [stat.me] 20 Dec 2012
arxiv:12125049v1 [statme] 20 Dec 2012 UNIVERSITÀ CATTOLICA DEL SACRO CUORE DIPARTIMENTO DI SCIENZE STATISTICHE A PARTIAL LEAST SQUARES ALGORITHM HANDLING ORDINAL VARIABLES ALSO IN PRESENCE OF A SMALL NUMBER
More informationUCLA STAT 233 Statistical Methods in Biomedical Imaging
UCLA STAT 233 Statistical Methods in Biomedical Imaging Instructor: Ivo Dinov, Asst. Prof. In Statistics and Neurology University of California, Los Angeles, Spring 2004 http://www.stat.ucla.edu/~dinov/
More informationReconciling conflicting evidence on the origins of comparative development: A finite mixture model approach
Reconciling conflicting evidence on the origins of comparative development: A finite mixture model approach Thomas K.J. Grantham Research Institute on Climate Change and the Environment, London School
More informationUSDA Dairy Import License Circular for 2018
USDA Dairy Import License Circular for 2018 Commodity/Note Country Name TRQ Jan Feb Mar Apr May Jun Grand Total Non-Cheese 21,864,781 624,064 651,121 432,669 901,074 1,202,567 907,493 4,718,988 BUTTER
More informationEXTENDING PARTIAL LEAST SQUARES REGRESSION
EXTENDING PARTIAL LEAST SQUARES REGRESSION ATHANASSIOS KONDYLIS UNIVERSITY OF NEUCHÂTEL 1 Outline Multivariate Calibration in Chemometrics PLS regression (PLSR) and the PLS1 algorithm PLS1 from a statistical
More informationOptimal design of experiments
Optimal design of experiments Session 4: Some theory Peter Goos / 40 Optimal design theory continuous or approximate optimal designs implicitly assume an infinitely large number of observations are available
More informationIntermediate Econometrics
Intermediate Econometrics Markus Haas LMU München Summer term 2011 15. Mai 2011 The Simple Linear Regression Model Considering variables x and y in a specific population (e.g., years of education and wage
More informationAbout the Authors Geography and Tourism: The Attraction of Place p. 1 The Elements of Geography p. 2 Themes of Geography p. 4 Location: The Where of
Preface p. ix About the Authors p. xi Geography and Tourism: The Attraction of Place p. 1 The Elements of Geography p. 2 Themes of Geography p. 4 Location: The Where of Geography p. 4 Place and Space:
More information2017 Source of Foreign Income Earned By Fund
2017 Source of Foreign Income Earned By Fund Putnam Emerging Markets Equity Fund EIN: 26-2670607 FYE: 08/31/2017 Statement Pursuant to 1.853-4: The fund is hereby electing to apply code section 853 for
More informationDYNAMIC AND COMPROMISE FACTOR ANALYSIS
DYNAMIC AND COMPROMISE FACTOR ANALYSIS Marianna Bolla Budapest University of Technology and Economics marib@math.bme.hu Many parts are joint work with Gy. Michaletzky, Loránd Eötvös University and G. Tusnády,
More informationEuroGeoSurveys & ASGMI The Geological Surveys of Europe and IberoAmerica
EuroGeoSurveys & ASGMI The Geological Surveys of Europe and IberoAmerica Geological Surveys, what role? Legal mandate for data & information: Research Collection Management Interpretation/transformation
More informationThe Colonial Origins of French Trade Patterns
The Colonial Origins of French Trade Patterns Tania Kallab Advisor: Cristina Terra 24 October, 2013 that I. Sommaire Research Question and Motivation Colonialism and Economic Performance (AJR 2001, Alam
More informationSelected Answers for Core Connections Algebra
Selected Answers for Core Connections Algebra Lesson 9.1.1 9-6. a: (w +14) 2 = 144 ; w = 2 or 26 b: (x + 2.5) 2 = 2.25; x = 1 or 4 c: (k! 8) 2 = 81 ; k = 1 or 17 d: (z! 500) 2 = 189225 ; z = 65 or 935
More informationDISCRIMINATION AND CLASSIFICATION IN NIR SPECTROSCOPY. 1 Dept. Chemistry, University of Rome La Sapienza, Rome, Italy
DISCRIMINATION AND CLASSIFICATION IN NIR SPECTROSCOPY Federico Marini Dept. Chemistry, University of Rome La Sapienza, Rome, Italy Classification To find a criterion to assign an object (sample) to one
More informationOn the Identification of a Class of Linear Models
On the Identification of a Class of Linear Models Jin Tian Department of Computer Science Iowa State University Ames, IA 50011 jtian@cs.iastate.edu Abstract This paper deals with the problem of identifying
More informationHigh-dimensional regression modeling
High-dimensional regression modeling David Causeur Department of Statistics and Computer Science Agrocampus Ouest IRMAR CNRS UMR 6625 http://www.agrocampus-ouest.fr/math/causeur/ Course objectives Making
More informationTMM UPDATE TRANS DAY OF REMEMBRANCE 2017
TMM UPDATE TRANS DAY OF REMEMBRANCE 2017 2,609 reported murders of trans and gender-diverse people between 1 January 2008 and 30 September 2017 World Regions/Countries All regions Africa South Africa 1
More informationUSDA Dairy Import License Circular for 2018 Commodity/
USDA Dairy Import License Circular for 2018 Commodity/ Grand Country Name TRQ Jan Feb Mar Apr May Jun Jul Aug Sep Note Total Non-Cheese 21,864,781 624,064 651,121 432,669 901,074 1,202,567 907,493 1,117,261
More informationPredicting unobserved links in incompletely observed networks
Computational Statistics & Data Analysis 52 (2008) 1373 1386 www.elsevier.com/locate/csda Predicting unobserved links in incompletely observed networks David J. Marchette a,, Carey E. Priebe b a Naval
More informationSituation on the death penalty in the world. UNGA Vote 2012 Resolutio n 67/176. UNGA Vote 2010 Resolutio n 65/206. UNGA Vote 2008 Resolutio n 63/168
Situation on the death penalty in the world Prepared by the International Commission against the Death Penalty (ICDP), as of 8 June 2014. Based on Amnesty International and Death Penalty Worldwide. Country
More informationOn the relationships among latent variables and residuals in PLS path modeling: The formative-reflective scheme
Computational Statistics & Data Analysis 51 (2007) 5828 5846 wwwelseviercom/locate/csda On the relationships among latent variables and residuals in PLS path modeling: The formative-reflective scheme Giorgio
More informationFactor Analysis. Qian-Li Xue
Factor Analysis Qian-Li Xue Biostatistics Program Harvard Catalyst The Harvard Clinical & Translational Science Center Short course, October 7, 06 Well-used latent variable models Latent variable scale
More information2 D wavelet analysis , 487
Index 2 2 D wavelet analysis... 263, 487 A Absolute distance to the model... 452 Aligned Vectors... 446 All data are needed... 19, 32 Alternating conditional expectations (ACE)... 375 Alternative to block
More informationInference using structural equations with latent variables
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike License. Your use of this material constitutes acceptance of that license and the conditions of use of materials on this
More information2001 Environmental Sustainability Index
2001 Environmental Sustainability Index Annex 6: Variable Descriptions and Data An Initiative of the Global Leaders of Tomorrow Environment Task Force, World Economic Forum Annual Meeting 2001 Davos, Switzerland
More informationICC Rev August 2010 Original: English. Agreement. International Coffee Council 105 th Session September 2010 London, England
ICC 105-7 Rev. 1 31 August 2010 Original: English Agreement E International Coffee Council 105 th Session 21 24 September 2010 London, England Obstacles to consumption Background 1. In accordance with
More information1. Impacts of Natural Disasters by Region, 2008
1. Impacts of Natural Disasters by Region, 2008 Among all regions across the world in 2008, Asia not only ranks first but also dominates in all natural disaster s impact categories occurrence, killed,
More informationContributions to partial least squares regression and supervised principal component analysis modeling
University of New Mexico UNM Digital Repository Mathematics & Statistics ETDs Electronic Theses and Dissertations 2-9-2010 Contributions to partial least squares regression and supervised principal component
More informationConjoint use of variables clustering and PLS structural equations modelling
Conjoint use of variables clustering and PLS structural equations modelling Valentina Stan 1 and Gilbert Saporta 1 1 Conservatoire National des Arts et Métiers, 9 Rue Saint Martin, F 75141 Paris Cedex
More informationCountry of Citizenship, College-Wide - All Students, Fall 2014
Country of Citizenship, College-Wide - All Students, Fall 2014-49,552 (72%) students were U.S. Citizens in Fall 2014. - MDC's non-citizen students come from at least 167 countries and speak approximately
More informationTHE OBJECTIVE FUNCTION OF PARTIAL LEAST SQUARES REGRESSION
THE OBJECTIVE FUNCTION OF PARTIAL LEAST SQUARES REGRESSION CAJO J. F. TER BRAAK Centre for Biometry Wageningen, P.O. Box 1, NL-67 AC Wageningen, The Netherlands AND SIJMEN DE JONG Unilever Research Laboratorium,
More informationII&Ij <Md Tmlaiiiiiit, aad once in Ihe y a w Teataa m i, the vmb thatalmta oot Uiaapirit world. into as abode or wotld by them- CooTBOtioa
382 4 7 q X
More informationNew York State Learning Standards and Core Curriculum Science Grade: 8 - Adopted: Structure and Properties of Matter
Main Criteria: New York State Learning Standards and Core Curriculum Secondary Criteria: Subjects: Science, Social Studies Grade: 8 Correlation Options: Show Correlated New York State Learning Standards
More informationReconciling factor-based and composite-based approaches to structural equation modeling
Reconciling factor-based and composite-based approaches to structural equation modeling Edward E. Rigdon (erigdon@gsu.edu) Modern Modeling Methods Conference May 20, 2015 Thesis: Arguments for factor-based
More informationUncertainty quantification and visualization for functional random variables
Uncertainty quantification and visualization for functional random variables MascotNum Workshop 2014 S. Nanty 1,3 C. Helbert 2 A. Marrel 1 N. Pérot 1 C. Prieur 3 1 CEA, DEN/DER/SESI/LSMR, F-13108, Saint-Paul-lez-Durance,
More informationFor more information about how to cite these materials visit
Author(s): Kerby Shedden, Ph.D., 2010 License: Unless otherwise noted, this material is made available under the terms of the Creative Commons Attribution Share Alike 3.0 License: http://creativecommons.org/licenses/by-sa/3.0/
More informationNeatest and Promptest Manner. E d i t u r ami rul)lihher. FOIt THE CIIILDIIES'. Trifles.
» ~ $ ) 7 x X ) / ( 8 2 X 39 ««x» ««! «! / x? \» «({? «» q «(? (?? x! «? 8? ( z x x q? ) «q q q ) x z x 69 7( X X ( 3»«! ( ~«x ««x ) (» «8 4 X «4 «4 «8 X «x «(» X) ()»» «X «97 X X X 4 ( 86) x) ( ) z z
More informationUsing Estimating Equations for Spatially Correlated A
Using Estimating Equations for Spatially Correlated Areal Data December 8, 2009 Introduction GEEs Spatial Estimating Equations Implementation Simulation Conclusion Typical Problem Assess the relationship
More informationBare minimum on matrix algebra. Psychology 588: Covariance structure and factor models
Bare minimum on matrix algebra Psychology 588: Covariance structure and factor models Matrix multiplication 2 Consider three notations for linear combinations y11 y1 m x11 x 1p b11 b 1m y y x x b b n1
More informationMeasurement Theory. Reliability. Error Sources. = XY r XX. r XY. r YY
Y -3 - -1 0 1 3 X Y -10-5 0 5 10 X Measurement Theory t & X 1 X X 3 X k Reliability e 1 e e 3 e k 1 The Big Picture Measurement error makes it difficult to identify the true patterns of relationships between
More informationOptimization. Charles J. Geyer School of Statistics University of Minnesota. Stat 8054 Lecture Notes
Optimization Charles J. Geyer School of Statistics University of Minnesota Stat 8054 Lecture Notes 1 One-Dimensional Optimization Look at a graph. Grid search. 2 One-Dimensional Zero Finding Zero finding
More informationOffice of Budget & Planning 311 Thomas Boyd Hall Baton Rouge, LA Telephone 225/ Fax 225/
Louisiana Acadia 20 17 3 0 0 0 Allen 2 2 0 0 0 0 Ascension 226 185 37 2 1 1 Assumption 16 15 1 0 0 0 Avoyelles 20 19 1 0 0 0 Beauregard 16 11 4 0 0 1 Bienville 2 2 0 0 0 0 Bossier 22 18 4 0 0 0 Caddo 91
More informationDemography, Time and Space
Demography, Time and Space Martin Bell The University of Queensland WD Borrie Lecture Australian Population Association 2014 Conference Hobart, Tasmania Wednesday December 3rd 2014 Professor WD (Mick)
More informationOffice of Budget & Planning 311 Thomas Boyd Hall Baton Rouge, LA Telephone 225/ Fax 225/
Louisiana Acadia 25 19 4 2 0 0 Allen 8 7 1 0 0 0 Ascension 173 143 26 1 0 3 Assumption 14 12 2 0 0 0 Avoyelles 51 41 9 0 0 1 Beauregard 18 14 3 0 0 1 Bienville 5 0 4 0 1 0 Bossier 28 27 0 1 0 0 Caddo 95
More informationUnconstrained Ordination
Unconstrained Ordination Sites Species A Species B Species C Species D Species E 1 0 (1) 5 (1) 1 (1) 10 (4) 10 (4) 2 2 (3) 8 (3) 4 (3) 12 (6) 20 (6) 3 8 (6) 20 (6) 10 (6) 1 (2) 3 (2) 4 4 (5) 11 (5) 8 (5)
More informationOn the Equivalence Between Canonical Correlation Analysis and Orthonormalized Partial Least Squares
On the Equivalence Between Canonical Correlation Analysis and Orthonormalized Partial Least Squares Liang Sun, Shuiwang Ji, Shipeng Yu, Jieping Ye Department of Computer Science and Engineering, Arizona
More informationI L L I N O I S UNIVERSITY OF ILLINOIS AT URBANA-CHAMPAIGN
Canonical Edps/Soc 584 and Psych 594 Applied Multivariate Statistics Carolyn J. Anderson Department of Educational Psychology I L L I N O I S UNIVERSITY OF ILLINOIS AT URBANA-CHAMPAIGN Canonical Slide
More informationGINA Children. II Global Index for humanitarian Needs Assessment (GINA 2004) Sheet N V V VI VIII IX X XI XII XII HDR2003 HDR 2003 UNDP
Human UNICEF Index Index Natural 2003 GDP per Total as % of Total Rate HDI HPI Disasters Conflicts capita Population population 5 1 Congo, Democratic Republic of the 2,80000 3 3 1 3 3 3 3 3 3 3 2 Burundi
More informationGIS Final Project: Stage of Development Index. country s stage of development and how this relates to their impact on nature and the
Group 4 Bridget Blair Mikhail Cunill Quinn Magevney Jordan Repollo GIS Final Project: Stage of Development Index Main Goal: Our goal is to take variables that are not inherently geospatial to better understand
More informationStochastic optimization - how to improve computational efficiency?
Stochastic optimization - how to improve computational efficiency? Christian Bucher Center of Mechanics and Structural Dynamics Vienna University of Technology & DYNARDO GmbH, Vienna Presentation at Czech
More informationEconometrics I. Professor William Greene Stern School of Business Department of Economics 3-1/29. Part 3: Least Squares Algebra
Econometrics I Professor William Greene Stern School of Business Department of Economics 3-1/29 Econometrics I Part 3 Least Squares Algebra 3-2/29 Vocabulary Some terms to be used in the discussion. Population
More informationState-space Model. Eduardo Rossi University of Pavia. November Rossi State-space Model Fin. Econometrics / 53
State-space Model Eduardo Rossi University of Pavia November 2014 Rossi State-space Model Fin. Econometrics - 2014 1 / 53 Outline 1 Motivation 2 Introduction 3 The Kalman filter 4 Forecast errors 5 State
More informationPartitioned Covariance Matrices and Partial Correlations. Proposition 1 Let the (p + q) (p + q) covariance matrix C > 0 be partitioned as C = C11 C 12
Partitioned Covariance Matrices and Partial Correlations Proposition 1 Let the (p + q (p + q covariance matrix C > 0 be partitioned as ( C11 C C = 12 C 21 C 22 Then the symmetric matrix C > 0 has the following
More informationConnectivity in the LAC region
Cristián Varas, Speedchecker Ltd. Agustín Formoso, Lacnic Labs Connectivity in the LAC region A study made by LACNIC Labs and Speedchecker Ltd. Outline The challenge of measuring regions not well covered
More informationIntroduction to time-use surveys
Regional seminar on time-use surveys Introduction to time-use surveys Kingstown, Saint Vincent and the Grenadines 10-11 December 2014 Iliana Vaca Lucía Scuro Division of Gender Affairs WHAT IS TIME? Measured
More informationTemperature and Income: Online Appendices
Temperature and Income: Online Appendices Melissa Dell Massachusetts Institute of Technology Benjamin F. Jones Northwestern University and NBER Benjamin A. Olken Massachusetts Institute of Technology and
More informationKernel Partial Least Squares for Nonlinear Regression and Discrimination
Kernel Partial Least Squares for Nonlinear Regression and Discrimination Roman Rosipal Abstract This paper summarizes recent results on applying the method of partial least squares (PLS) in a reproducing
More informationIntroduction to Factor Analysis
to Factor Analysis Lecture 10 August 2, 2011 Advanced Multivariate Statistical Methods ICPSR Summer Session #2 Lecture #10-8/3/2011 Slide 1 of 55 Today s Lecture Factor Analysis Today s Lecture Exploratory
More informationState-space Model. Eduardo Rossi University of Pavia. November Rossi State-space Model Financial Econometrics / 49
State-space Model Eduardo Rossi University of Pavia November 2013 Rossi State-space Model Financial Econometrics - 2013 1 / 49 Outline 1 Introduction 2 The Kalman filter 3 Forecast errors 4 State smoothing
More informationCanonical Correlation Analysis with Kernels
Canonical Correlation Analysis with Kernels Florian Markowetz Max-Planck-Institute for Molecular Genetics Computational Molecular Biology Berlin Computational Diagnostics Group Seminar 2003 Mar 10 1 Overview
More informationMore formally, the Gini coefficient is defined as. with p(y) = F Y (y) and where GL(p, F ) the Generalized Lorenz ordinate of F Y is ( )
Fortin Econ 56 3. Measurement The theoretical literature on income inequality has developed sophisticated measures (e.g. Gini coefficient) on inequality according to some desirable properties such as decomposability
More informationFrom dummy regression to prior probabilities in PLS-DA
JOURNAL OF CHEMOMETRICS J. Chemometrics (2007) Published online in Wiley InterScience (www.interscience.wiley.com).1061 From dummy regression to prior probabilities in PLS-DA Ulf G. Indahl 1,3, Harald
More informationChapter 10 Conjoint Use of Variables Clustering and PLS Structural Equations Modeling
Chapter 10 Conjoint Use of Variables Clustering and PLS Structural Equations Modeling Valentina Stan and Gilbert Saporta Abstract In PLS approach, it is frequently assumed that the blocks of variables
More informationLearning with Singular Vectors
Learning with Singular Vectors CIS 520 Lecture 30 October 2015 Barry Slaff Based on: CIS 520 Wiki Materials Slides by Jia Li (PSU) Works cited throughout Overview Linear regression: Given X, Y find w:
More informationANALYSIS OF NONLINEAR PARTIAL LEAST SQUARES ALGORITHMS
ANALYSIS OF NONLINEAR PARIAL LEAS SQUARES ALGORIHMS S. Kumar U. Kruger,1 E. B. Martin, and A. J. Morris Centre of Process Analytics and Process echnology, University of Newcastle, NE1 7RU, U.K. Intelligent
More informationFORMATIVE AND REFLECTIVE MODELS: STATE OF THE ART. Anna Simonetto *
Electronic Journal of Applied Statistical Analysis EJASA (2012), Electron. J. App. Stat. Anal., Vol. 5, Issue 3, 452 457 e-issn 2070-5948, DOI 10.1285/i20705948v5n3p452 2012 Università del Salento http://siba-ese.unile.it/index.php/ejasa/index
More informationRNR 516A. Computer Cartography. Spring GIS Portfolio
RNR 516A Computer Cartography Spring 2016 GIS Portfolio 1 Contents 1 Political and Locator Maps 3 2 Base Maps and Digitizing 4 3 Data Entry Report 5 4 Projections and Symbolization 6 5 Choropleth Mapping
More informationLecture notes in Generalized Linear Models
Lecture notes in Generalized Linear Models Germán Rodríguez Princeton University http://data.princeton.edu/wws509 August 19, 2013 This page is added by Håvard Rue 1 Chapter 2 Linear Models for Continuous
More informationSTATISTICAL LEARNING SYSTEMS
STATISTICAL LEARNING SYSTEMS LECTURE 8: UNSUPERVISED LEARNING: FINDING STRUCTURE IN DATA Institute of Computer Science, Polish Academy of Sciences Ph. D. Program 2013/2014 Principal Component Analysis
More informationHUDSONVILLE MIDDLE SCHOOL COURSE FRAMEWORK
HUDSONVILLE MIDDLE SCHOOL COURSE FRAMEWORK COURSE/SUBJECT 7 th Grade Social Studies UNIT PACING Unit 1 ~ Geography of the Western World: Europe, Latin America, The United States and Canada September 6
More informationWe are IntechOpen, the world s leading publisher of Open Access books Built by scientists, for scientists. International authors and editors
We are IntechOpen, the world s leading publisher of Open Access books Built by scientists, for scientists 3,500 108,000 1.7 M Open access books available International authors and editors Downloads Our
More informationResponse Surface Methodology III
LECTURE 7 Response Surface Methodology III 1. Canonical Form of Response Surface Models To examine the estimated regression model we have several choices. First, we could plot response contours. Remember
More informationTutorial: Causal Model Search
Tutorial: Causal Model Search Richard Scheines Carnegie Mellon University Peter Spirtes, Clark Glymour, Joe Ramsey, others 1 Goals 1) Convey rudiments of graphical causal models 2) Basic working knowledge
More informationDISTILLED SPIRITS - IMPORTS BY VALUE DECEMBER 2017
DISTILLED SPIRITS - IMPORTS BY VALUE DECEMBER 2017 U.S. DUTIABLE IMPORTS OF DISTILLED SPIRITS (U.S. DOLLARS) Ea-Value-17-12 SUMMARY: IMPORTS ENTERED FOR CONSUMPTION CLASS DECEMBER DECEMBER PERCENT JANUARY
More information5 th Grade Social Studies Goals for the First Trimester Miss Gaull
5 th Grade Social Studies Goals for the First Trimester Overview: American People, American Land The American People Government by the People Free Enterprise Lands and Regions Resources and the Environment
More information