output: html_document: theme: readable —

Figure 1:  QQ plots of weightMZ (left), heightMZ (middle) and BMIMZ (right)
among 2-19-year-olds in NHANES.  The black points represent values high BIVs
and the triangles, low BIVs; the grey points represent data for values that
were considered to be biologically plausible based on the WHO cut-points.  
If the data were normally distributed, all points would fall on a diagonal
line.  In contrast, both weight and BMI were skewed to the right with very
heavy tails.  The distribution of height was much closer to that of a normal
distribution.  

Figure 1

f2 +  geom_point(data=dslb,cex=.z,shape=.s,color=.c) +
  geom_point(data=dslw,cex=.z,shape=.s,color=.c) + 
  geom_point(data=dslh,cex=.z,shape=.s, color=.c)

Figure 2:  Lines represent the weighted, smoothed 85th, 95th and 99th
percentiles of waist circumference (top), arm circumference (middle), and leg
length (bottom) by sex (columns) and age for all children in NHANES 1999-2012
as calculated with quantile regression  (35,36). The points in the top and
middle panels represent the waist and arm circumferences of the 113 children
who had a BMIMZ between 6 and 6.9, the triangles represent BMIMZ values = 7.0.
The points in the bottom panel represent the leg lengths of the 63 children 
who had a heightMZ = 3.0.

Figure 2

f1 + geom_point(data=dmp1, color='black', 
                   aes(y=value, shape=1), size=2) +
     geom_point(data=dmp2, color='black', 
                aes(y=value, shape=17), size=2) +
     scale_shape_identity() + 
     labs(x='\nAge (years)', y='Value (cm)\n', linetype=.leg, shape=.leg,
          color=.leg)