Inference for (multiscale) geographically weighted regression

MgwrLocalT: local t-values with the da Silva-Fotheringham corrected critical value \(t_{1-\alpha/(2 ENP), df}\). BandwidthConfidenceInterval: interval of candidate bandwidths carrying the requested cumulative Akaike weight (Li et al. 2020). Identical to the Python arm morie.fn.mgwrinfer.

Usage

MgwrLocalT(coef, se, enp, alpha = 0.05, df = NULL)

BandwidthConfidenceInterval(bandwidths, aicc, level = 0.95)

Arguments

coef, se: Local coefficients and standard errors.
enp: Effective number of parameters of the surface.
alpha: Family-wise level.
df: Degrees of freedom (default n - enp).
bandwidths, aicc: Candidate bandwidths and their AICc.
level: Cumulative Akaike weight.

Value

List.

References

da Silva, A. R. and Fotheringham, A. S. (2016). The multiple testing issue in geographically weighted regression. Geographical Analysis 48, 233-247.

Li, Z., Fotheringham, A. S., Oshan, T. M. and Wolf, L. J. (2020). Measuring bandwidth uncertainty in multiscale geographically weighted regression using Akaike weights. Annals of the American Association of Geographers 110, 1500-1520.

Examples

MgwrLocalT(c(0.5, -0.2, 1.1), c(0.2, 0.25, 0.3), 2, df = 50)$significant
#> [1]  TRUE FALSE  TRUE
BandwidthConfidenceInterval(c(40, 50, 60, 70, 80), c(310, 302, 300, 301, 306))
#> $best
#> [1] 60
#> 
#> $lower
#> [1] 50
#> 
#> $upper
#> [1] 70
#> 
#> $weights
#> [1] 0.003317657 0.181137959 0.492384022 0.298646005 0.024514357
#> 
#> $coverage
#> [1] 0.972168
#>