prediction and plotting cascades

1 files changed, 17 insertions, 18 deletions

diff --git a/R Scripts/predict-victims-plots.R b/R Scripts/predict-victims-plots.R
index 0655d28..dee3cdc 100644
--- a/R Scripts/predict-victims-plots.R
+++ b/R Scripts/predict-victims-plots.R
@@ -1,27 +1,32 @@
 ##### Plot results
+# load('Results/correct_rank_91415.RData')
 
+nvics = dim(correct_rank)[1]
 correct_rank1 = correct_rank[,length(lambdas)] # demographics model
 correct_rank2 = correct_rank[,1] # cascade model
 correct_rank3 = correct_rank[,which.min(colMeans(correct_rank))] # best combined model
-counts = matrix(c( sum(correct_rank1<(vcount(lcc)*0.0005)),
-                  sum(correct_rank1<(vcount(lcc)*0.001)),
-                  sum(correct_rank1<(vcount(lcc)*0.01)),
-                  sum(correct_rank2<(vcount(lcc)*0.0005)),
-                  sum(correct_rank2<(vcount(lcc)*0.001)),
-                  sum(correct_rank2<(vcount(lcc)*0.01)),
-                  sum(correct_rank3<(vcount(lcc)*0.0005)),
-                  sum(correct_rank3<(vcount(lcc)*0.001)),
-                  sum(correct_rank3<(vcount(lcc)*0.01))),
+popsizes = c(0.1,0.5,1.0)/100
+vcount(lcc)*popsizes
+counts = matrix(c( sum(correct_rank1<(vcount(lcc)*popsizes[1])),
+                  sum(correct_rank1<(vcount(lcc)*popsizes[2])),
+                  sum(correct_rank1<(vcount(lcc)*popsizes[3])),
+                  sum(correct_rank2<(vcount(lcc)*popsizes[1])),
+                  sum(correct_rank2<(vcount(lcc)*popsizes[2])),
+                  sum(correct_rank2<(vcount(lcc)*popsizes[3])),
+                  sum(correct_rank3<(vcount(lcc)*popsizes[1])),
+                  sum(correct_rank3<(vcount(lcc)*popsizes[2])),
+                  sum(correct_rank3<(vcount(lcc)*popsizes[3]))),
                 nrow=3, byrow=T)
 counts = counts*100/nvics
 barplot(counts, 
         xlab="Size of High-Risk Population",
         ylab="Percent of Victims Predicted",
-        names.arg=c('0.1%','0.5%','1%'),ylim=c(0,max(counts)*1.1),
+        names.arg=paste(as.character(popsizes*100),'%',sep=''),
+        ylim=c(0,max(counts)*1.1),
         col=c(rgb(0,0,1,1/2),rgb(1,0,0,1/2),rgb(0,1,0,1/2)),
         beside=TRUE)
 legend("topleft", inset=0.05, 
-       c("Demographics (Logit)", "Cascades (Sum parent)", "Combined Model"), 
+       c("Demographics", "Cascades", "Combined Model"), 
        fill=c(rgb(0,0,1,1/2),rgb(1,0,0,1/2),rgb(0,1,0,1/2)))
 box(which='plot')
 
@@ -34,18 +39,12 @@ barplot(counts,
 axis(side = 4)
 mtext(side = 4, line = 3, "Number of Victims Predicted")
 
-popsizes = c(0.1, 0.5, 1)
-plot(popsizes,counts[1,],type='l',ylim=c(0,max(counts)))
-lines(popsizes,counts[2,])
-lines(popsizes,counts[3,])
-lines(c(0,1),c(0,1))
 
 #### Precision-Recall Curve
-plot(ecdf(correct_rank1),col='red',xlim=c(0,50),lwd=2)
+plot(ecdf(correct_rank1),col='red',lwd=2,xlim=c(1,100))
 plot(ecdf(correct_rank2),col='darkblue',lwd=2,add=T)
 plot(ecdf(correct_rank3),col='darkgreen',lwd=2,add=T)
 legend("bottomright", inset=0.05, 
        c("Demographics Model", "Cascade Model", "Combined Model"), 
        fill=c('red','darkblue','darkgreen'))
-lines(c(0,vcount(lcc)),c(0,1))