parameters out of range: ets, optim

Question

I have written a code in R which adds weighting and runs additive holt-winters to forecast. However for some of my data it gives the error:

Error in etsmodel(y, errortype[i], trendtype[j], seasontype[k], damped[l], : Parameters out of range

Can someone please tell me why it is doing this and how i can stop it happening in future.

Here is my code:

suppressMessages(library(lmtest))
suppressMessages(library(car))
suppressMessages(library(tseries))
suppressMessages(library(forecast))
suppressMessages(library(TTR))
suppressMessages(library(geoR))
suppressMessages(library(MASS))

#-------------------------------------------------------------------------------
Input.data <- matrix(c("08Q1","08Q2","08Q3","08Q4","09Q1","09Q2","09Q3","09Q4","10Q1","10Q2","10Q3","10Q4","11Q1","11Q2","11Q3","11Q4","12Q1","12Q2","12Q3","12Q4","13Q1","13Q2","13Q3","13Q4","14Q1","14Q2","14Q3",73831.11865,84750.47149,85034.80061,99137.19637,62626.50672,72144.77761,74726.1774,122203.5416,84872.02354,96054.77537,93849.93456,136380.3862,94252.32737,101044.518,112453.256,138807.2089,102091.1436,102568.8303,98839.36528,129249.4421,91207.28917,93060.79801,87776.30512,124342.2055,87128.55797,90261.46195,86371.5614),ncol=2,byrow=FALSE)


Frequency <- 1/4

Forecast.horizon <- 4

Start.date <- c(8, 1)

Data.col <- as.numeric(Input.data[, length(Input.data[1, ])])

Data.col.ts <- ts(Data.col, deltat=Frequency, start = Start.date)

trans<- abs(round(BoxCox.lambda(Data.col, method = "loglik"),5))
categ<-as.character( c(cut(trans,c(0,0.25,0.75,Inf),right=FALSE)) )
Data.new<-switch(categ,
                 "1"=log(Data.col.ts),
                 "2"=sqrt(Data.col.ts),
                 "3"=Data.col.ts
)

mape <- function(percent.error)              
  mean(abs(percent.error))
#----- Weighting ---------------------------------------------------------------
fweight <- function(x){
  PatX <- 0.5+x 
  return(PatX)
}

integvals <- rep(0, length.out = length(Data.new))
for (i in 1:length(Data.new)){
  integi <- integrate(fweight, lower = (i-1)/length(Data.new), upper= i/length(Data.new))
  integvals[i] <- 2*integi$value
}

HWAW <- ets(Data.new, model = "AAA", damped = FALSE, opt.crit = "mse", ic="aic", lower = c(0.03, 0.03, 0.03, 0.04), 
            upper = c(0.997, 0.997, 0.997, 0.997), bounds = "usual", restrict = FALSE)
parASW <- round(HWAW$par[1:3], digits=3)
HWAOPT <- function(parASW)
{
  HWAddW <- ets(Data.new, model = "AAA", alpha = parASW[1], beta = parASW[2], gamma = parASW[3], damped = FALSE, opt.crit = "mae", ic="aic",
                lower = c(0.001, 0.001, 0.001, 0.0001), upper = c(0.999, 0.999, 0.999, 0.999), bounds = "admissible", restrict = FALSE)
  error <- c(resid(HWAddW))
  error <- t(error) %*% integvals
  percent.error <- 100*(error/c(Data.new))
  MAPE <- mape(percent.error)
  return(MAPE)
}
OPTHWA <- optim(parASW, HWAOPT, method="L-BFGS-B", lower=c(rep(0.01, 3)), upper=c(rep(0.99, 3)), control = list(fnscale= 1, maxit = 3000))
# Alternatively, set  method="Nelder-Mead" or method="L-BFGS-B" 
parS4 <- OPTHWA$par
HWAW1 <- ets(Data.new, model = "AAA", alpha = parS4[1], beta = parS4[2], gamma = parS4[3], damped = FALSE, opt.crit = "mae", ic="aic",
             lower = c(0, 0, 0, 0), upper = c(0.999, 0.999, 0.999, 0.999), bounds = "admissible", restrict = FALSE)

Thankyou in advance

Edit:

Even when removing the limits for the upper and lower bound the error remains

Edit

I removed the opt.crit from the ets which made my code run fine. If there is another way then please let me know

EDIT

Although this worked for this data set, it still gave an error for a different one. So there must be something else i can do to make this code run automatically for all data sets

I've taken out the bits that arent needed, but everything else is essential — Summer-Jade Gleek'away, Nov 28 '14 at 12:26
It doesn't look like you have done any basic debugging. If you don't know how to do that, search the web and Stack Overflow or ask a question about debugging practices eventually. Find out where exactly the error occurs and check if the input to the function call that throws the error is what you expect it to be. — Roland, Nov 28 '14 at 12:34
i have debugged and it says the error comes from when it checks the parameters and lower and upper bounds, but when i removed the bounds it still gave the error — Summer-Jade Gleek'away, Nov 28 '14 at 12:36

score 1 · Answer 1 · answered Oct 03 '18 at 20:11

Most likely the parameters estimated by HWAW which you then round to 3 decimal points in parS4 are inadmissible. Beta must be smaller than alpha and gamma must be smaller than 1 minus alpha:

https://github.com/robjhyndman/forecast/issues/179

https://robjhyndman.com/eindhoven/2-1-StateSpace.pdf

Check the parameters you are passing in:

HWAW1 <- ets(..., alpha = parS4[1], beta = parS4[2], gamma = parS4[3], ...)

score 0 · Answer 2 · edited Nov 04 '19 at 10:39

I am writing a hw model with alpha, beta and gamma parameters with beta less than alpha and gamma less than beta. I am still getting the error.
Here is my model and error message.

tsx <- ts(data = x, start = c(2016, 10), frequency = 12)
hw(y = tsx, h = 16, alpha = 0.6, beta = 0.3, gamma = 0.2)

The error message I get

Error in etsmodel(y, errortype[i], trendtype[j], seasontype[k], damped[l], : Parameters out of range

parameters out of range: ets, optim

2 Answers2