Description Usage Arguments Details Value Author(s) See Also Examples
Constrain unlimited growth phase of a single respiration time series.
1 2 | kinrespGrowthphaseReplicate(rder, weights = NULL,
orderTime = TRUE, residType = "pearson")
|
rder |
data.frame with columns replicate, time, and resp |
weights |
Weighting of the observations for non-equal precision |
orderTime |
if rder is already ordered by time, then you can |
residType |
type of residuals, see argument type of |
It discardsa all data after the inflection point
If the curve deviates from the exponential model, residuals will be correlated.
The longest time series is selected for which there is no correlation or a negative correlation determined by Breusch-Godfrey Test (bgtest) and Durbin-Watson-Test (dwtest)
list of class kinresp with components
the subset with exponential growth phase
the data of the entire growth phase, i.e. until maximum respiration rate
the gnls fitting object
the number of points suggested by different criteria. Entry 1 (named "n") gives the best combined estimate.
the complete statistics r2 and p-values of various residual tests
results of all the fits. Used e.g. for plotting diagnostics
Thomas Wutzler <thomas.wutzler@web.de>
kinrespGrowthphaseExperiment
,plotKinrespDiagnostics.kinresp
,twKinresp
1 2 3 4 5 6 7 8 9 10 11 12 13 | # we pick and plot the respiration time series of Fig 1 in Wutzler et al. 2010
# data(respWutzler10)
rder <- subset(
respWutzler10, suite == "Face" & experiment == 3 & replicate == 2 )
plot( resp ~ time, data = rder )
res2 <- kinrespGrowthphaseReplicate(rder, weights = varPower(fixed = 0.5))
res2$n["n"] #display the number of records
#display the fitting line
lines( fitted(res2$fit) ~ getUnlimitedGrowthData(res2)$time )
# plot diagnostics
plotKinrespDiagnostics(res2) #use arrow keys to go back in plot history
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.