PRINCIPAL COMPONENT ANALYSIS(PCA) EOFs and Principle Components; Selection Rules LECTURE 8 Supplementary Readings: Wilks, chapters 9.

Slides:



Advertisements
Similar presentations
FMRI Methods Lecture 10 – Using natural stimuli. Reductionism Reducing complex things into simpler components Explaining the whole as a sum of its parts.
Advertisements

Principal Component Analysis (PCA) or Empirical Orthogonal Functions (EOFs) Arnaud Czaja (SPAT Data analysis lecture Nov. 2011)
Principal Component Analysis for SPAT PG course Joanna D. Haigh.
Noise & Data Reduction. Paired Sample t Test Data Transformation - Overview From Covariance Matrix to PCA and Dimension Reduction Fourier Analysis - Spectrum.
Chapter 28 – Part II Matrix Operations. Gaussian elimination Gaussian elimination LU factorization LU factorization Gaussian elimination with partial.
Face Recognition Ying Wu Electrical and Computer Engineering Northwestern University, Evanston, IL
Lecture 18 Varimax Factors and Empircal Orthogonal Functions.
Environmental Data Analysis with MatLab Lecture 16: Orthogonal Functions.
Lecture 7: Principal component analysis (PCA)
Principal Component Analysis CMPUT 466/551 Nilanjan Ray.
Principal Component Analysis
Linear Transformations
Unsupervised Learning - PCA The neural approach->PCA; SVD; kernel PCA Hertz chapter 8 Presentation based on Touretzky + various additions.
The rank of a product of two matrices X and Y is equal to the smallest of the rank of X and Y: Rank (X Y) =min (rank (X), rank (Y)) A = C S.
Slide 1 EE3J2 Data Mining EE3J2 Data Mining Lecture 9(b) Principal Components Analysis Martin Russell.
Face Recognition Using Eigenfaces
The Terms that You Have to Know! Basis, Linear independent, Orthogonal Column space, Row space, Rank Linear combination Linear transformation Inner product.
6 1 Linear Transformations. 6 2 Hopfield Network Questions.
Lecture 20 Empirical Orthogonal Functions and Factor Analysis.
Duan Wang Center for Polymer Studies, Boston University Advisor: H. Eugene Stanley.
What is EOF analysis? EOF = Empirical Orthogonal Function Method of finding structures (or patterns) that explain maximum variance in (e.g.) 2D (space-time)
Slide 1 EE3J2 Data Mining EE3J2 Data Mining Lecture 9 Data Analysis Martin Russell.
Techniques for studying correlation and covariance structure
Chapter 2 Dimensionality Reduction. Linear Methods
Presented By Wanchen Lu 2/25/2013
Next. A Big Thanks Again Prof. Jason Bohland Quantitative Neuroscience Laboratory Boston University.
Recap of PCA: what it does, how to do it Details of PCA presentation of results terminology scaling truncation of PCs interpretation of PCs Rotation of.
Extensions of PCA and Related Tools
D. van Alphen1 ECE 455 – Lecture 12 Orthogonal Matrices Singular Value Decomposition (SVD) SVD for Image Compression Recall: If vectors a and b have the.
El Niño-Southern Oscillation in Tropical Column Ozone and A 3.5-year signal in Mid-Latitude Column Ozone Jingqian Wang, 1* Steven Pawson, 2 Baijun Tian,
What is it? Principal Component Analysis (PCA) is a standard tool in multivariate analysis for examining multidimensional data To reveal patterns between.
Canonical Correlation Analysis and Related Techniques Simon Mason International Research Institute for Climate Prediction The Earth Institute of Columbia.
N– variate Gaussian. Some important characteristics: 1)The pdf of n jointly Gaussian R.V.’s is completely described by means, variances and covariances.
SINGULAR VALUE DECOMPOSITION (SVD)
Slicing and dicing the 153-year record of monthly sea level at San Francisco, California using singular spectrum analysis Larry Breaker Moss Landing Marine.
Correlation of temperature with solar activity (SSN) Alexey Poyda and Mikhail Zhizhin Geophysical Center & Space Research Institute, Russian Academy of.
Central limit theorem revisited
Principal Components: A Mathematical Introduction Simon Mason International Research Institute for Climate Prediction The Earth Institute of Columbia University.
MULTIVARIATE REGRESSION Multivariate Regression; Selection Rules LECTURE 6 Supplementary Readings: Wilks, chapters 6; Bevington, P.R., Robinson, D.K.,
Introduction to Linear Algebra Mark Goldman Emily Mackevicius.
Christina Bonfanti University of Miami- RSMAS MPO 524.
EIGENSYSTEMS, SVD, PCA Big Data Seminar, Dedi Gadot, December 14 th, 2014.
Principle Component Analysis and its use in MA clustering Lecture 12.
Principal Component Analysis (PCA)
Feature Extraction 主講人:虞台文. Content Principal Component Analysis (PCA) PCA Calculation — for Fewer-Sample Case Factor Analysis Fisher’s Linear Discriminant.
3 “Products” of Principle Component Analysis
REGRESSION (CONTINUED) Matrices & Matrix Algebra; Multivariate Regression LECTURE 5 Supplementary Readings: Wilks, chapters 6; Bevington, P.R., Robinson,
Université d’Ottawa / University of Ottawa 2001 Bio 8100s Applied Multivariate Biostatistics L11.1 Lecture 11: Canonical correlation analysis (CANCOR)
Feature Extraction 主講人:虞台文.
Oceanography 569 Oceanographic Data Analysis Laboratory Kathie Kelly Applied Physics Laboratory 515 Ben Hall IR Bldg class web site: faculty.washington.edu/kellyapl/classes/ocean569_.
Chapter 13 Discrete Image Transforms
Central limit theorem revisited Throw a dice twelve times- the distribution of values is not Gaussian Dice Value Number Of Occurrences.
Central limit theorem - go to web applet. Correlation maps vs. regression maps PNA is a time series of fluctuations in 500 mb heights PNA = 0.25 *
Unsupervised Learning II Feature Extraction
EMPIRICAL ORTHOGONAL FUNCTIONS 2 different modes SabrinaKrista Gisselle Lauren.
Slicing and dicing the 153-year record of monthly
Spatial Modes of Salinity and Temperature Comparison with PDO index
9.3 Filtered delay embeddings
Factor analysis Advanced Quantitative Research Methods
Lecture: Face Recognition and Feature Reduction
Principal Component Analysis
Singular Value Decomposition North Atlantic SST
Principal Component Analysis
Recitation: SVD and dimensionality reduction
Singular Value Decomposition SVD
X.1 Principal component analysis
Outline Singular Value Decomposition Example of PCA: Eigenfaces.
Principal Component Analysis
Canonical Correlation Analysis and Related Techniques
Marios Mattheakis and Pavlos Protopapas
Presentation transcript:

PRINCIPAL COMPONENT ANALYSIS(PCA) EOFs and Principle Components; Selection Rules LECTURE 8 Supplementary Readings: Wilks, chapters 9

WE’LL START OUT WITH AN EXAMPLE: 20th GLOBAL SURFACE TEMPERATURE RECORD

Climatic Research Unit (‘CRU’), University of East Anglia Surface Temperature Changes

EOFs for the five leading eigenvectors of the global temperature data from The gridpoint areal weighting factor used in the PCA procedure has been removed from the EOFs so that relative temperature anomalies can be inferred from the patterns. 12% (88%) 6% (3%) 5% (1%) 4% (1%) 3% (0.5%) EOF #1 EOF #2 EOF #3 EOF #4 EOF #5

SURFACE TEMPERATURE RECORD FILTERED BY RETAINING PROJECTION ONTO WITH FIRST FIVE EIGENVECTORS FILTERING THROUGH PCA

GLOBAL TEMPERATURE TREND EOF #1 PC #1

Multivariate ENSO Index (“MEI”) EOF #2 PC #2 EL NINO/SOUTHERN OSCILLATION (ENSO)

NORTH ATLANTIC OSCILLATION EOF #3 PC #3

EOF #3 PC #3 NORTH ATLANTIC OSCILLATION

TROPICAL ATLANTIC “DIPOLE” EOF #3 PC #3

ATLANTIC MULTIDECADAL OSCILLATION EOF #5 PC #5

EOF #5 PC #5 ATLANTIC MULTIDECADAL OSCILLATION

EOF #5 PC #5 ATLANTIC MULTIDECADAL OSCILLATION

PCA as an SVD on the Data Matrix X

Recall from our earlier lecture the variance-covariance matrix A in the multivariate regression problem: The eigenvectors of A comprise an orthogonal predictor set (Principal Components Regression)

Let us return to the data matrix, (assume it has zero mean) We can write Where U,V are unitary matrices (orthogonal matrices if X is real-valued), U is MxN, S is diagonal NxN, and V is NxN Singular Value Decomposition (SVD) Assume M>N (overdetermined; greater number of “equations” than “unknowns”)

We can then write Where U, V are unitary matrices (orthogonal matrices if X is real-valued), U is NxM, S is diagonal MxM, and V is MxM Singular Value Decomposition (SVD) Typically, we are interested in the case N>M. A revised overdetermined problem can be obtained by redefining the problem:

V is a unitary matrix which diagonalizes XX T ! There is a mathematical equivalence between taking the Singular Value Decomposition (SVD) of X, and finding the eigenvectors of A=XX T Thus, S 2 contains the eigenvalues of XX T

U contains as its columns the temporal patterns or Principal Components (“PC”s) corresponding to the M eigenvalues, which are the “right eigenvectors” of the SVD: V contains the as its columns the Spatial Pattern or Empirical Orthogonal Function (“EOF”) corrresponding to the M eigenvalues, which are the “left eigenvectors” of the SVD:

We can filter the original data with a subset of M* eigenvectors: FILTERING WITH EIGENVECTORS

Standardization & Areal Weighting Gappy Data Frequency domain “Rotation” Selection Rules Some Additional Considerations:

How many eigenvectors do we consider significant? Eigenvalue > 1/M Break in slope in eigenvalue spectrum (“Scree” test) or log eigenvalue (“LEV”) spectrum Eigenvalue lies outside expected distribution for M uncorrelated Gaussian time series of length N (Preisendorfer Rule N). This is an example of a Monte Carlo method Rule N’ (take into account serial correlation) There is no uniquely defensible criterion... SELECTION RULES

Preisdendorfer Rule N SELECTION RULES

Asymptotic results of Preisendorfer Rule N for large sample size (N,M>100 or so) =N/M=N/M SELECTION RULES

MATLAB EXAMPLE: NORTH ATLANTIC SEA LEVEL PRESSURE DATA