melanoma.csv
proc import datafile="F:\publish\data analysis using R\data\melanoma.csv" out=melanoma dbms=csv replace;
run;
proc genmod data=melanoma;
class age region / param=ref;
model cases = age region / dist=poisson link=log offset=ltotal type3;
run;
Standard Wald 95% Confidence Wald
Parameter DF Estimate Error Limits Chi-Square Pr > ChiSq
Intercept 1 -9.8388 0.0974 -10.0297 -9.6480 10211.4 <.0001
age 35-44 1 1.7974 0.1209 1.5604 2.0344 220.92 <.0001
age 45-54 1 1.9131 0.1184 1.6810 2.1452 260.90 <.0001
age 55-64 1 2.2418 0.1183 2.0099 2.4737 358.89 <.0001
age 65-74 1 2.3657 0.1315 2.1080 2.6235 323.56 <.0001
age 75+ 1 2.9447 0.1320 2.6859 3.2035 497.30 <.0001
region north 1 -0.8195 0.0710 -0.9587 -0.6803 133.11 <.0001
Scale 0 1.0000 0.0000 1.0000 1.0000
NOTE: The scale parameter was held fixed.
LR Statistics For Type 3 Analysis
Chi-
Source DF Square Pr > ChiSq
age 5 796.74 <.0001
region 1 124.22 <.0001
proc nlmixed data=melanoma;
parms mu=-10, alpha1=2, alpha2=2, alpha3=2, alpha4=2, alpha5=2, beta=1;
eta=mu+beta*(region="north")+alpha1*(age="35-44")+alpha2*(age="45-54")
+alpha3*(age="55-64")+alpha4*(age="65-74")+alpha5*(age="75+");
lambda=exp(eta)*total;
model cases ~ poisson(lambda);
run;
Standard
Parameter Estimate Error DF t Value Pr > |t| Alpha Lower Upper Gradient
mu -9.8388 0.09736 12 -101.05 <.0001 0.05 -10.0510 -9.6267 0.000085
alpha1 1.7974 0.1209 12 14.86 <.0001 0.05 1.5339 2.0609 -0.00001
alpha2 1.9131 0.1184 12 16.15 <.0001 0.05 1.6550 2.1711 0.000011
alpha3 2.2418 0.1183 12 18.94 <.0001 0.05 1.9840 2.4996 -0.00007
alpha4 2.3657 0.1315 12 17.99 <.0001 0.05 2.0792 2.6523 -0.00006
alpha5 2.9447 0.1320 12 22.30 <.0001 0.05 2.6570 3.2324 0.000202
beta -0.8195 0.07103 12 -11.54 <.0001 0.05 -0.9742 -0.6647 0.00013