mgwr.gwr.GWRResults¶

class
mgwr.gwr.
GWRResults
(model, params, predy, S, CCT, influ, tr_STS=None, w=None)[source]¶ Basic class including common properties for all GWR regression models
 Parameters
 modelGWR object
pointer to GWR object with estimation parameters
 paramsarray
n*k, estimated coefficients
 predyarray
n*1, predicted y values
 Sarray
n*n, hat matrix
 CCTarray
n*k, scaled variancecovariance matrix
 warray
n*1, final weight used for iteratively reweighted least sqaures; default is None
 Attributes
 modelGWR Object
points to GWR object for which parameters have been estimated
 paramsarray
n*k, parameter estimates
 predyarray
n*1, predicted value of y
 yarray
n*1, dependent variable
 Xarray
n*k, independent variable, including constant
 familyfamily object
underlying probability model; provides distributionspecific calculations
 ninteger
number of observations
 kinteger
number of independent variables
 df_modelinteger
model degrees of freedom
 df_residinteger
residual degrees of freedom
 offsetarray
n*1, the offset variable at the ith location. For Poisson model this term is often the size of the population at risk or the expected size of the outcome in spatial epidemiology; Default is None where Ni becomes 1.0 for all locations
 scalefloat
sigma squared used for subsequent computations
 warray
n*1, final weights from iteratively reweighted least sqaures routine
 resid_responsearray
n*1, residuals of the repsonse
 resid_ssscalar
residual sum of sqaures
 Warray
n*n; spatial weights for each observation from each calibration point
 Sarray
n*n, hat matrix
 CCTarray
n*k, scaled variancecovariance matrix
ENP
scalareffective number of parameters
tr_S
floattrace of S (hat) matrix
 tr_STSfloat
trace of STS matrix
y_bar
arrayweighted mean of y
TSS
arraygeographically weighted total sum of squares
RSS
arraygeographically weighted residual sum of squares
R2
floatGlobal rsquared value for a Gaussian model.
adj_R2
floatAdjusted global rsquared for a Gaussian model.
 aicfloat
Akaike information criterion
 aiccfloat
corrected Akaike information criterion to account to account for model complexity (smaller bandwidths)
 bicfloat
Bayesian information criterio
localR2
arraylocal R square
sigma2
floatresidual variance
std_res
arraystandardized residuals
bse
arraystandard errors of Betas
 influarray
n*1, leading diagonal of S matrix
 CooksDarray
n*1, Cook’s D
tvalues
arrayReturn the tstatistic for a given parameter estimate.
adj_alpha
arrayCorrected alpha (critical) values to account for multiple testing during hypothesis testing.
 deviancearray
n*1, local model deviance for each calibration point
 resid_deviancearray
n*1, local sum of residual deviance for each calibration point
 llfscalar
loglikelihood of the full model; see pysal.contrib.glm.family for damilysepcific loglikelihoods
pDev
floatLocal percentage of deviance accounted for.
D2
floatPercentage of deviance explanied.
adj_D2
floatAdjusted percentage of deviance explanied.
 muarray
n*, flat one dimensional array of predicted mean response value from estimator
 fit_paramsdict
parameters passed into fit method to define estimation routine
 predictionsarray
p*1, predicted values generated by calling the GWR predict method to predict dependent variable at unsampled points ()
Methods
D2
(self)Percentage of deviance explanied.
ENP
(self)effective number of parameters
R2
(self)Global rsquared value for a Gaussian model.
RSS
(self)geographically weighted residual sum of squares
TSS
(self)geographically weighted total sum of squares
adj_D2
(self)Adjusted percentage of deviance explanied.
adj_R2
(self)Adjusted global rsquared for a Gaussian model.
adj_alpha
(self)Corrected alpha (critical) values to account for multiple testing during hypothesis testing.
bse
(self)standard errors of Betas
conf_int
(self)Returns the confidence interval of the fitted parameters.
cooksD
(self)Influence: leading diagonal of S Matrix
cov_params
(self, cov[, exog_scale])Returns scaled covariance parameters
critical_tval
(self[, alpha])Utility function to derive the critical tvalue based on given alpha that are needed for hypothesis testing
filter_tvals
(self[, critical_t, alpha])Utility function to set tvalues with an absolute value smaller than the absolute value of the alpha (critical) value to 0.
localR2
(self)local R square
local_collinearity
(self)Computes several indicators of multicollinearity within a geographically weighted design matrix, including:
pDev
(self)Local percentage of deviance accounted for.
sigma2
(self)residual variance
spatial_variability
(self, selector[, …])Method to compute a Monte Carlo test of spatial variability for each estimated coefficient surface.
std_res
(self)standardized residuals
summary
(self)Print out GWR summary
tr_S
(self)trace of S (hat) matrix
tvalues
(self)Return the tstatistic for a given parameter estimate.
use_t
(self)bool(x) > bool
y_bar
(self)weighted mean of y
W
adj_pseudoR2
aic
aicc
bic
deviance
df_model
df_resid
global_deviance
initialize
llf
llnull
normalized_cov_params
null
null_deviance
pearson_chi2
predictions
pseudoR2
pvalues
resid_anscombe
resid_deviance
resid_pearson
resid_response
resid_ss
resid_working
scale