Spatial Survey Data Modeling - Sampling Spatial Units for Agricultural Surveys - page 317

> framepop < - cbind(framepop,probinc)
> set.seed(160964)
> pps < - UPtille(probinc)
> pps < - round(pps)
> framepps < - framepop[pps ¼¼ 1,]
> dpps < - svydesign (id ¼ ~1,data ¼ framepps,fpc ¼ ~framepps$probinc,
+ pps ¼ "brewer")
> epps < - svyglm(yobs ~ poly(xc,2)+poly(yc,2),dpps)
> summary(epps)

Call:
svyglm(formula ¼ yobs ~ poly(xc, 2) + poly(yc, 2), dpps)

Survey design:
svydesign(id ¼ ~1, data ¼ framepps, fpc ¼ ~framepps$probinc,
pps ¼ "brewer")

Coefficients:
Estimate Std. Error t value Pr( > |t|)
(Intercept) 114.7624 0.5171 221.943 < 2e-16 ***

poly(xc, 2)1 1.3039 4.7346 0.275 0.7836

poly(xc, 2)2 -76.0830 3.1144 -24.429 < 2e-16 ***

poly(yc, 2)1 9.1787 3.2981 2.783 0.0065 **

poly(yc, 2)2 -43.0014 4.2352 -10.153 < 2e-16 ***

---
Signif. codes: 0 ' *** ' 0.001 ' ** ' 0.01 ' * ' 0.05 ' . ' 0.1 '' 1

(Dispersion parameter for gaussian family taken to be 13.64098)

Number of Fisher Scoring iterations: 2

> regpps < - lm(yobs ~ poly(xc,2)+poly(yc,2), data ¼ framepps)
> summary(regpps)

Call:
lm(formula ¼ yobs ~ poly(xc, 2) + poly(yc, 2), data ¼ framepps)

Residuals:
Min 1Q Median 3Q Max
-10.0199 -3.0870 -0.1716 3.2992 13.1509
Coefficients:
Estimate Std. Error t value Pr( > |t|)
(Intercept) 116.9027 0.4644 251.720 < 2e-16 ***

poly(xc, 2)1 2.4476 4.7694 0.513 0.609006

poly(xc, 2)2 -74.6390 4.6604 -16.016 < 2e-16 ***