similar to: HoltWinters problem

Hi R-users, I have a question relating to the HoltWinters() function. I am trying to forecast a series using the Holt Winters methodology but I am getting some unusual results. I had previously been using R for Windows version 2.7.2 and have just started using R 2.9.1. While using version 2.7.2 I was getting reasonable results however upon changing versions I found I started to see unusual

Full_Name: John Bodley Version: 2.5.1 (2007-06-27) OS: Windows XP Submission from: (NULL) (12.144.182.66) I was fitting a number of time series in R using the stats::HoltWinters method to define a single exponential smoothing model, i.e., beta = gamma = 0. I came across an example where the fitted value of alpha was not defined in the [0, 1] interval which seems to violate the lower and upper

I get John's value (48.8789) in 2.7.0 and R-devel (both on Ubuntu). Really seems to be a numeric issue: > HoltWinters(x, beta = 0, gamma = 0)$alpha alpha 48.87989 > HoltWinters(x * 1.0000000001, beta = 0, gamma = 0)$alpha alpha 0.6881547 > HoltWinters(x * 1.00000000001, beta = 0, gamma = 0)$alpha alpha 48.87989 Providing starting values seems to help, but not

Dear helpers I'm having a problem with function HoltWinters from package ts. I have a time series that I want to fit an Holt-Winters model and make predictions for the next values. I've already built an object of class ts to serve as input to HoltWinters. But then I get an error; I've used HoltWinters a lot of times and this never hapened > data.HW<-HoltWinters(data.ts) Error

An update on this: I just patched HoltWinters() to use optimize() in the univariate case, and it now computes the correct value. David John Bodley wrote: > Hi, > > Thanks for the quick response. I upgraded by version of R on Windows to the > latest (2.7.0) and re-ran the analysis and get the same result of 48.87989. > > The original time series was a non-regular zoo()

Hi. I have to forecast a time series of a Internet network traffic bitrate. The data are in file http://www.forumaltavilla.it/joomla/datitesi/dati.datand the sampling time is every 0.05 seconds. Now, i want to use HoltWinters forecasting. This is my script. dt=1.58443823e-9 #0.05 seconds in years dati.ts=ts(scan("dati.dat"),start=0,deltat=dt) model=HoltWinters(dati.ts)

Hi everyone, I'm using the HoltWinters() function to do a time series analysis. The function only returns the back fitted values ($fitted) after the first year of data, which is my case, is a little more than half. However, when I use the plot() function, it plots the back fit for almost the entire data set. Any ideas on how to extract the fitted values going all the way back to the start

Hey All, First time on these forums. Thanks in advance. Soooo... I have a process that was functioning well before the 2.14 update. Now the HoltWinters function is throwing an error whereby I get the following: Error in HoltWinters(sales.ts) : optimization failure I've been looking around to determine why this happens (see if I can test the data beforehand) but I haven't come

It doesn't do it on my system (I get a value of about 0.688 in R 2.7.0 patched on Linux), and 2.5.1 is not current. Does a better starting value help? However, HoltWinters is using optim() in a case it is not designed for (one-dimensional optimization): see the note on its help page. I think this could easily be changed, but as HoltWinters is contributed code I am Cc:ing the author for

every time i try to run HoltWinters i get this error message: > HoltWinters(z, seasonal="additive") Error in decompose(ts(x[1:wind], start = start(x), frequency = f), seasonal) : time series has no or less than 3 periods what's going on? somebody please help me. -- View this message in context:

Hello all, I am looking for some help in understanding how to change the way R optimizes the smoothing constant selection process for the HoltWinters function. I'm a SAS veteran but very new to R and still learning my way around. Here is some sample data and the current HoltWinters code I'm using: rawdata <- c(294, 316, 427, 487, 441, 395, 473, 423, 389, 422, 458, 411, 433, 454,

Hello, I would like to look at the compiled C code behind HoltWinters from the stats package. Is that possible? If so where do I find it? thanks, Spencer [[alternative HTML version deleted]]

I want to forecaste the call number everyday for a call-center. Now I have removed the influence of the fluctuation with some method, so only thing left is to analyze the trend of the call number every day. I have thought of two ways: regression and HoltWinters smooth. But when I use regression, I find some day's call number will bcome negative, which is obviously unreasonabe. If I use

Hi, I am trying to fit the HoltWinters exponential smoothing on a monthly time series data in R. My questions are: 1. I know that the level, trend and seasonality are updated over time. So are the output coefficients a, b and s1-s12 for a specific time t (a,b and s12 for last observataion for example) ? 2. I am trying to work out the fitted values but I could not figure out the h value used in

Dear list members I want to analyze separately the months of a time series. In other words, I want to plot and fit models for each month separately. Taking the example of http://a-little-book-of-r-for-time-series.readthedocs.org/en/latest/src/timeseries.html births <- scan("http://robjhyndman.com/tsdldata/data/nybirths.dat") birthstimeseries <- ts(births, frequency=12,

I have a set of data (in a matrix). I spliced a column out and parsed it as.ts (time series). I then plotted the time series but I found that it was very noisy. I wanted to smooth it out. However, I am having some problems smoothing and plotting the smoothed version. > A <- as.ts(read.table(choose.files())) > x <- as.ts(A[,10]) > plot(x) > > plot(smooth(x))

Dear All, I was wondering why the forecast for an additive HoltWinters model is given by Yhat[t+h] = a[t] + h * b[t] + s[t + 1 + (h - 1) mod p]. I am a student and new to time series analysis and forecasting. That said, I considered t = 13 and h = 1: Yhat[13+1] = a[13] + b[13] + s[13 + 1] It seems odd that to predict Yhat[14], you would need a s[14] which in turn depends on Y[14], given that

Hello everyone, I was wondering if anyone was aware of a way in which I could use ggplot's stat_smooth() function for add an exponential moving average. I was thinking that I could maybe use something like: >myggplot + stat_smooth (method = 'HoltWinters( data , .9 , 0, 0)') but my efforts were futile. Perhaps there is a way to write my own custom method to throw into

Hi, i want to create an object which contains different classes: for example i have some time series and test if ARIMA models are best than HoltWinters models: for each of my time serie i want to collect in an unique object which model was the best: for some it will be an HoltWinters class and for some other an Arima class. is there any solution?? thanks in advance, with my best, erik sauleau

Hi, I've noted that not all `predict' methods are hidden in the namespace: > methods("predict") [1] predict.ar* predict.Arima* [3] predict.arima0* predict.glm [5] predict.HoltWinters* predict.lm [7] predict.loess* predict.mlm [9] predict.nls* predict.poly [11] predict.ppr* predict.prcomp* [13]