centrend {NADA2}

Trend analysis of censored data with a covariate

Description

Computes the ATS (Mann-Kendall trend test for censored data) after adjustment of censored data for a covariate.

Usage

centrend(
  y.var,
  y.cens,
  x.var,
  time.var,
  link = "identity",
  Smooth = "cs",
  printstat = TRUE,
  stackplots = FALSE
)

Arguments

y.var

The column of y (response variable) values plus detection limits

y.cens

The column of indicators, where 1 (or TRUE) indicates a detectionlimit in the y.var column, and 0 (or FALSE) indicates a detected value in y.var.

x.var

Column of a covariate (not time). y.var will be smoothed versus x.var and residuals taken to subtract out the relationship between y and x.

time.var

Column of the time variable, either a sequence of days or decimal times, etc. Will be the scale used for time in ATS trend analysis.

link

Default = ⁠“identity”⁠ which means it uses data in the original units. See details.

Smooth

Type of smoother used in the GAM. Default is ⁠“cs”⁠, shrinkage cubic regression splines. See details for other options.

printstat

Logical TRUE/FALSE option of whether to print the resulting statistics in the console window, or not. Default is TRUE.

stackplots

logical 'TRUE'/'FALSE' option to stack three plots that are output onto the same page instead of each on separate page. Default is 'FALSE', each separate.

Details

Default link = identity. The y variables are then used in their original units. Other options are available see cenGAM::tobit1 for more options.

Default Smooth is "cs" for shrinkage cubic regression splines. See mgcv::smooth.terms for other types of smoothing algorithms. '"ts"' is a thin-plate regression spline and is also commonly used.

Value

Prints three plots: Y data vs time with GAM Smooth, Residuals from GAM Smooth vs time, and ATS trend line of residuals vs time.

Returns GAM residuals and ATS results on trend test of residuals (intercept, slope, Kendall's tau, p-value for trend)

References

Helsel, D.R., 2011. Statistics for censored environmental data using Minitab and R, 2nd ed. John Wiley & Sons, USA, N.J.

See Also

mgcv::gam

Examples

data(Brumbaugh)

Brumbaugh$time=1:nrow(Brumbaugh)

with(Brumbaugh,centrend(Hg,HgCen,SedTotHg,time.var=time))