optimise.ERP: optimise_ERP

In connolr3/eegdatasim: Tool for EEG signals and ERPs

optimise_ERP

Description

optimise uses OPTIM package to find parameters of ERP that minimise the SSE... Usually, using OPTIM in R is equivalemt to fitting a model to a time series. However our model is based on trigonemtric functions.

optimise_ERP uses min_sse function as the function to minimise. It uses BFGS method to optimise. The gradient of the SSE is supplied also, which speeds up computation (gradient is the derivative of the SSE function).

Altough optime_ERP estimate both amplitude and frequency..... the optimisation is only one-dimensional as the amplitude can be formed in terms of the freqeuncy.

Usage

optimise_ERP(dat,startingfrequency=0,starting_amp=1,mysr,pkcntr)

Arguments

x	x values from data we are fitting
y	y values from data we are fitting
mysr	Sampling Rate, in Hertz, used and knwon by experimenter
pkcntr	Index of peak center, if unknown please estimate with the which(max(dat))

Value

par	The best set of parameters found. par[1] = freq, par[2]=amplitude
value	The value of fn corresponding to par.
counts	A two-element integer vector giving the number of calls to fn and gr respectively. This excludes those calls needed to compute the Hessian, if requested, and any calls to fn to compute a finite-difference approximation to the gradient.
convergence	The value of fn corresponding to par. An integer code. 0 indicates successful completion (which is always the case for "SANN" and "Brent"). Possible error codes are 1 indicates that the iteration limit maxit had been reached. 10 indicates degeneracy of the Nelder–Mead simplex. 51 indicates a warning from the "L-BFGS-B" method; see component message for further details. 52 indicates an error from the "L-BFGS-B" method; see component message for further details.
message	Any additional info returned by optimiser, may be Null

Note

Please see: https://www.rdocumentation.org/packages/stats/versions/3.6.2/topics/optim for R documentation on Optim Function

Author(s)

Rose Connolly [e-mail]

Examples

# mysignal<-noise(200,10,250)+3*peak(200,10,250,7,115)
# av_signal<-signal_averaging1(mysignal,200,10)
# hats<-est_sig_hat(av_signal)
# stan_signal<-(av_signal-hats[2])/hats[1]
# erp_is<-find_ERP_range(stan_signal,cutoff=1.6)
#
#
# peak_center_estimate=which(av_signal==max(av_signal))
# pars<-optimise_ERP(erp_is,av_signal[erp_is],mysr=250,peak_center_estimate)
#
#
# print(paste("estimating frequency as:", pars$par[1]))
# print(paste("Actual Frequency was:", 7))
# (paste("estimating amplitude as",pars$par[2]))
# print(paste("Actual Amplitude:", 3))


## The function is currently defined as
gr_min_SSE <- function(par,x,y,srate,peakcenter){
  u <- ((x-peakcenter)*2*pi)/srate
  z <- cos( u * par[1] )
  alphaest <- sum( z * y) / sum(z*z)
  ans<-4*pi*alphaest*(x-peakcenter)*(y-alphaest*z)*(sin(u * par[1]))
  return(sum(ans/srate))
}

min_SSE<-function(par,x,y,srate,peakcenter){
  z <- cos(((x-peakcenter)*2*pi*par[1])/srate)
  alphaest <- sum(z * y) / sum(z*z)
  #pred_values<-cos(((data["x"]-peakcenter)*2*pi*par[1])/srate)
  errs<- y - alphaest * z #pred_values
  sum(errs^2)
}

#' @export
optimise_ERP<-function(x,y,mysr,pkcntr){
  result <- optim(par = 1, fn = min_SSE, gr=gr_min_SSE, x=x, y=y,srate=mysr,peakcenter=pkcntr, method="BFGS")
  z <- cos(((x-pkcntr)*2*pi*result$par[1])/mysr )
  alphaest <- sum(z * y) / sum(z*z)
  result$par <- c((result$par),alphaest)
  return(result)
}

connolr3/eegdatasim documentation built on April 14, 2022, 10:39 a.m.