GWR_WhitePaper - GEOGRAPHICALLY WEIGHTED REGRESSION WHITE...

Info iconThis preview shows pages 1–6. Sign up to view the full content.

View Full Document Right Arrow Icon
GEOGRAPHICALLY WEIGHTED REGRESSION WHITE PAPER MARTIN CHARLTON A STEWART FOTHERINGHAM National Centre for Geocomputation National University of Ireland Maynooth Maynooth, Co Kildare, IRELAND March 3 2009
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
2 The authors gratefully acknowledge support from a Strategic Research Cluster grant (07/SRC/I1168) by Science Foundation Ireland under the National Development Plan.
Background image of page 2
Contents Regression 1 Regression with Spatial Data 3 Geographically Weighted Regression (GWR) 5 Outputs from GWR 9 Interpreting Parameter Estimates 10 Extensions to GWR 12 References 13
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
1 Regression Regression encompasses a wide range of methods for modelling the relationship between a dependent variable and a set of one or more independent variables. The dependent variable is sometimes known as the y-variable, the response variable or the regressand. The independent variables are sometimes known as x-variables, predictor variables, or regressors. A regression model is expressed as an equation. In its simplest form a linear regression model can take the form i i i x y ε β + + = 1 0 for i=1 … n In this equation i y is the response variable, here measured at some location i , x i is the independent variable, i is the error term, and 0 and 1 are parameters 1 which are to be estimated such that the value of ( ) = n i i i y y 1 2 ˆ is minimised over the n observations in the dataset. The i y ˆ is the predicted or fitted value for the i th observation, given the i th value of x. The term ) ˆ ( i i y y is known as the residual for the i th observation, and the residuals should be both independent and drawn identically from a Normal Distribution with a mean of zero. Such a model is usually fitted using a procedure known as Ordinary Least Squares (OLS). More generally, a multiple linear regression model may be written: i mi m i i i x x x y + + + + + = ... 2 2 1 1 0 for i=1 … n where the predictions of the dependent variable are obtained through a linear combination of the independent variables. The OLS estimator takes the form: y X X X T T 1 ) ( ˆ = where ˆ is the vector of estimated parameters, X is the design matrix which contains the values of the independent variables and a column of 1s, y is the vector of observed values, and 1 ) ( X X T is the inverse of the variance-covariance matrix. 1 The term coefficient is sometimes used instead of parameter.
Background image of page 4
2 Sometimes it is desirable to weight the observations in a regression, for example, different levels of data uncertainty. The weights are placed in the leading diagonal of a square matrix W and the estimator is altered to include the weighting: Wy X WX X T T 1 ) ( ˆ = β The ability of the model to replicate the observed y values is measured by the goodness of fit. This is conveniently expressed by the r
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 6
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 02/15/2012 for the course GEO 4167 taught by Professor Staff during the Spring '12 term at University of Florida.

Page1 / 17

GWR_WhitePaper - GEOGRAPHICALLY WEIGHTED REGRESSION WHITE...

This preview shows document pages 1 - 6. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online