It's like a Histogram, but smoothed out

- The smooting is done with Kernels
- So it's an estimation of the Probability Density Function of the underlying variable

hist(pData$AGEP, breaks=12, col="red", prob=T) dens = density(pData$AGEP, adjust=2) lines(dens, col="blue", lwd=2)

dens = density(pData$AGEP) plot(dens, lwd=3, col="blue")

- this plot can also be used with two variables
- it is more convenient than Histograms for doing that
- since it uses plot, not bars

densM = density(pData$AGEP[which(pData$SEX==1)]) densF = density(pData$AGEP[which(pData$SEX==2)]) plot(densM, lwd=3, col="blue") lines(densF, lwd=3, col="orange")