Growth of soybean varieties in 3 years

Usage

data("davidian.soybean")

Format

A data frame with 412 observations on the following 5 variables.

plot: plot code
variety: variety, F or P
year: 1988-1990
day: days after planting
weight: weight of soybean leaves

Details

This experiment compared the growth patterns of two genotypes of soybean varieties: F=Forrest (commercial variety) and P=Plant Introduction number 416937 (experimental variety).

Data were collected in 3 consecutive years.

At the start of each growing season, 16 plots were seeded (8 for each variety). Data were collected approximately weekly. At each timepoint, six plants were randomly selected from each plot. The leaves from these 6 plants were weighed, and average leaf weight per plant was reported. (We assume that the data collection is destructive and different plants are sampled at each date).

Note: this data is the same as the "nlme::Soybean" data.

Source

Marie Davidian and D. M. Giltinan, (1995). Nonlinear Models for Repeated Measurement Data. Chapman and Hall, London.

Electronic version retrieved from https://www4.stat.ncsu.edu/~davidian/data/soybean.dat

References

Pinheiro, J. C. and Bates, D. M. (2000). Mixed-Effects Models in S and S-PLUS. Springer, New York.

Examples

if (FALSE) { # \dontrun{

  library(agridat)
  data(davidian.soybean)
  dat <- davidian.soybean
  dat$year <- factor(dat$year)

  libs(lattice)
  xyplot(weight ~ day|variety*year, dat,
         group=plot, type='l',
         main="davidian.soybean")


  # The only way to keep your sanity with nlme is to use groupedData objects
  # Well, maybe not.  When I use "devtools::run_examples",
  # the "groupedData" function creates a dataframe with/within(?) an
  # environment, and then "nlsList" cannot find datg, even though
  # ls() shows datg is visible and head(datg) is fine.
  # Also works fine in interactive mode. It is driving me insane.
  # reid.grasses has the same problem
  # Use if(0){} to block this code from running.
  if(0){
    libs(nlme)
    datg <- groupedData(weight ~ day|plot, dat)
    # separate fixed-effect model for each plot
    # 1988P6 gives unusual estimates
    m1 <- nlsList(SSlogis, data=datg,
                  subset = plot != "1988P6")
    # plot(m1) # seems heterogeneous
    plot(intervals(m1), layout=c(3,1)) # clear year,variety effects in Asym

    # A = maximum, B = time of half A = steepness of curve
    # C = sharpness of curve (smaller = sharper curve)

    # switch to mixed effects
    m2 <- nlme(weight ~ A / (1+exp(-(day-B)/C)),
               data=datg,
               fixed=list(A ~ 1, B ~ 1, C ~ 1),
               random = A +B +C ~ 1,
               start=list(fixed = c(17,52,7.5))) # no list!

    # add covariates for A,B,C effects, correlation, weights
    # not necessarily best model, but it shows the syntax
    m3 <- nlme(weight ~ A / (1+exp(-(day-B)/C)),
               data=datg,
               fixed=list(A ~ variety + year,
                          B ~ year,
                        C ~ year),
               random = A +B +C ~ 1,
               start=list(fixed= c(19,0,0,0,
                                   55,0,0,
                                   8,0,0)),
               correlation = corAR1(form = ~ 1|plot),
               weights=varPower(), # really helps
               control=list(mxMaxIter=200))

    plot(augPred(m3), layout=c(8,6),
      main="davidian.soybean - model 3")
  } # end if(0)
} # }