fitMod {seawaveQ}

Internal function that fits the seawaveQ model.

Description

fitMod is called from within fitswavecav but can be invoked directly. It fits the seawaveQ model and returns the results.

Usage

fitMod(
  cdatsub,
  cavdat,
  yrstart,
  yrend,
  tndbeg,
  tndend,
  tanm,
  pnames,
  qwcols,
  mclass = 1,
  numk = 4,
  plotfile = FALSE,
  textfile = FALSE
)

Arguments

cdatsub

is the concentration data.

cavdat

is the continuous (daily) ancillary data.

yrstart

is the starting year of the analysis (treated as January 1 of that year).

yrend

is the ending year of the analysis (treated as December 31 of that year).

tndbeg

is the beginning (in whole or decimal years) of the trend period.

tndend

is the end (in whole or decimal years) of the trend period.

tanm

is a character identifier that names the trend analysis run. It is used to label output files.

pnames

is the parameter (water-quality constituents) to analyze (if using USGS parameters, omit the starting 'P', such as "00945" for sulfate).

qwcols

is a character vector with the beginning of the column headers for remarks code (default is R), and beginning of column headers for concentration data (default is P for parameter).

mclass

indicates the class of model to use. A class 1 model is the the traditional SEAWAVE-Q model that has a linear time trend. A class 2 model is a newer option for longer trend periods that uses a set of restricted cubic splines on the time variable to provide a more flexible model. The default is 1. (Harrell, 2010, 2018).

numk

is the number of knots in the restricted cubic spline model (mclass = 2). The default is 4, and the recommended number is 3–7.

plotfile

is by default FALSE. True will write pdf files of plots to the user's file system.

textfile

is by default FALSE. True will write text output files to the user's file system. These files are useful for detailed model comparisons, documenting session information, and for model archives.

Value

A PDF file (if plotfile is TRUE) containing plots (see seawaveQPlots), a text file showing the best model survival regression call and results, and a list. The first element of the list contains information about the data and the model(s) selected (see examplestpars). The second element of the list contains the summary of the survival regression call. The third element of the list is itself a list containing the observed concentrations (censored and uncensored) and the predicted concentrations used by seawaveQPlots or seawaveQPlots2 to generate the plots.

Author(s)

Aldo V. Vecchia and Karen R. Ryberg

References

Allison, P.D., 1995, Survival analysis using the SAS system—A practical guide: Cary, N.C., SAS Institute, Inc., 304 p.

Harrell, F.E., Jr., 2010, Regression modeling strategies—With applications to linear models, logistic regression, and survival analysis: New York, Springer-Verlag, 568 p.

Harrell, F.E., Jr., 2018, rms—Regression modeling strategies: R package version 5.1-2, https://CRAN.R-project.org/package=rms.

Examples

data(swData)
myRes <- fitMod(cdatsub = examplecdatsub, cavdat = examplecavdat, 
yrstart = 1995, yrend = 2003, tndbeg = 1995, tndend = 2003, tanm = "myfit3", 
pnames = c("04041"), qwcols = c("R", "P"), plotfile = FALSE,
textfile = FALSE)