Cosine model fitting

This function performs the updated nonlinear least squares (NLS) regression method for the cosine model (see Chazin et al. 2019).

sineFit(data, amplitude = NULL, intercept = NULL, method = c("OLS", "initial"))

Arguments

data	A data frame that contains the data for one individual. There should be two columns with names 'distance' and 'oxygen'.
amplitude	Initial value for the amplitude parameter for the method="initial" method.
intercept	Initial value for the intercept parameter for the method="initial" method.
method	A character string giving the initialization for the nonlinear least squares regression. This must be either method="initial" or method="OLS". Default is method="OLS" method. method="initial" performs the nonlinear least squares (NLS) regression method for the cosine model without initializing parameter selections. It begins with the given initial values for amplitude and intercept. method="OLS" uses the least squares estimates (see Chazin et al. 2019) as the initial parameter selection.

Value

A fitted model object from the nls function in R:

m

an 'nlsModel' object incorporating the model.

convInfo

a list with convergence information

data

the expression that was passed to 'nls' as the data argument. The actual data values are present in the environment of the 'm' component.

call

the matched call with several components, notably 'algorithm'

dataClasses

the '"dataClasses"' attribute (if any) of the '"terms"' attribute of the model frame.

control

the control 'list' used

References

Florent Baty, Christian Ritz, Sandrine Charles, Martin Brutsche, Jean-Pierre Flandrois, Marie-Laure Delignette-Muller (2015). A Toolbox for Nonlinear Regression in R: The Package nlstools. Journal of Statistical Software, 66(5), 1-21. URL http://www.jstatsoft.org/v66/i05/.

Examples

armenia_split = split(armenia,f = armenia$ID)
amp = seq(1,10,by=0.5)
int = seq(-25,0,by=0.5)
sineFit(armenia_split[[2]],amp[3],int[4],method = "initial")
#> Nonlinear regression model
#>   model: oxygen ~ intercept + amplitude * cos(frequency * distance + phase)
#>    data: data
#> intercept amplitude     phase frequency 
#>   -5.7143    5.9191   -0.1394    0.1776 
#>  residual sum-of-squares: 1.049
#> 
#> Number of iterations to convergence: 8 
#> Achieved convergence tolerance: 1.426e-06
sineFit(armenia_split[[1]],method = "OLS")
#> Nonlinear regression model
#>   model: oxygen ~ intercept + amplitude * cos(frequency * distance + phase)
#>    data: data
#> intercept amplitude     phase frequency 
#>   -4.2485    5.2653    0.7701    0.1657 
#>  residual sum-of-squares: 1.371
#> 
#> Number of iterations to convergence: 6 
#> Achieved convergence tolerance: 2.407e-06