[R] How does one set up logical functions?

[Gabor Grothendieck]

On Tue, May 7, 2013 at 10:02 AM, Neotropical bat risk assessments
<neotropical.bats at gmail.com> wrote:
> Hi all,
>
> I am trying to set up logical function(s) to deal with two adjustments
> to a blood glucose value.
> I have been dinking around in Excel and assume this will be much easier
> in R.
>
> DF is date-time, BG value in mg/dL,test strip
> 4/3/13 19:20    105     Aviva-491350
> 4/4/13 21:03    74      Aviva-491350
> 4/6/13 17:40    81      Aviva-491640
> 4/6/13 17:40    82      Aviva-491350
> 4/6/13 22:48    106     Aviva-491640
> 4/6/13 22:48    102     Aviva-491350
> 4/7/13 5:32     87      Aviva-491350
> 4/7/13 5:32     103     Aviva-491640
>
>
> What I need are the high and low ranges based on "acceptable" standards
> of the measured values.
>
> The logical expressions need to be
> IF BG =>100 then "High limit" would = (BG+(BG*.15))
> IF BG =>100 then "Low limit" would = (BG-(BG*.15))
> and
> IF BG <100 then "High limit" would = (BG+15)
> IF BG <100 then "Low limit" would = (BG-15)
>
> The standards are written as: 95% of the individual glucose results
> shall fall within ą15 mg/dL of the reference results at glucose
> concentrations less than 100 mg/dL and within ą15% at glucose
> concentrations greater than or equal to 100 mg/dL.
>
> Then I need to plot the measured value and also show the high & low
> "acceptable" values.
>

Here it is using gglot2:


library(ggplot2)
library(gridExtra)

DF2 <- transform(DF,
datetime = as.POSIXct(DF2[[1]], format = "%m/%d/%y %H:%M"),
lower = ifelse(BG < 100, BG - 15, BG * 0.85),
upper = ifelse(BG < 100, BG + 15, BG * 1.15))


ggplot(DF2, aes(datetime, BG)) +
   geom_point() +
   geom_line() +
   geom_smooth(aes(ymin = lower, ymax = upper), stat = "identity") +
   geom_linerange(aes(ymin = lower, ymax = upper)) +
   annotation_custom(tableGrob(DF2, gp = gpar(cex = 0.5)), ymin = 120) +
   coord_cartesian(ylim = c(60, 150)) +
   xlab("") +
   ylab("Blood Glucose") +
   ggtitle("Blood Glucose Levels")


--
Statistics & Software Consulting
GKX Group, GKX Associates Inc.
tel: 1-877-GKX-GROUP
email: ggrothendieck at gmail.com