similar to: RHmm, mixture of gaussians, memory could not be "read" error

Hi, I am having difficulties estimating the parameters of a HMM using the HMM package. I have simulated a sequence of observations from a known HMM. When I estimate the parameters of a HMM using these simulated observations the parameters are not at all close to the known ones. I realise the estimated parameters are not going to be exactly the same as the known/true parameters, but these are

Dear all, I'm totally new to R. Recently I've been trying to use getYahooData in TTR package in order to download stock index daily open/high/low/close. The downloaded data is in the format of Open High Low Close Volume 2000-01-04 18937.45 19187.61 18937.45 19002.86 0 2000-01-05 19003.51 19003.51 18221.82 18542.55 0 2000-01-06

Hi r-help, I have been using your RHmm package for some time and have recently had to try using the package for a new dataset. Basically I have a dataset with a number of discrete observation variables that change over time, and I would love to try modeling them using a HMM. Basically I was wondering if RHmm can be used to model a multivariate discrete HMM, i.e., the observations are a vector

I am using getYahooData from TTR to get daily data. When i do it standalone, it is fine. It also works fine inside my code. However, when i run code inside mtrace(), i always get the following error: Error in xts(cbind(adj[[1]], adj[[2]]), index(obj)): order.by requires an appropriate time-based object After a lot of hand wringing, it looks to me that it happens because index in xts masks

Dear List, We are using RHmm to cluster data through HMM. We would like to restrict the transition matrix of HMM, to get hierarchical connections between clusters. But, RHmm doesn't seem to support these restrictions. Can any one suggest a library to do that. Thanks, Kishor

I admire package-builders being proactive and having their facilities ready for upcoming R releases. However, if the publicly released version of R is 2.10.1 and a package is built for R 2.11.0, users get the embarrassing notice about a disconnect, and cannot do much about it. If it is tedious to roll out a new package on time, perhaps there's a possibility for automation here. Thanks. I

library(xts) library(TTR) ndx = getYahooData("^NDX") aa = ndx$Close bb = aggregate(aa, as.yearweek, tail, 1) The last operation takes forever, and then the bb dates are messed up. The following produces the desired result: time(aa) = as.Date(time(aa)) bb = aggregate(aa, as.yearweek, tail, 1) The index of ndx and aa is of POSIXct (as reported by is(time(ndx))) , which apparently

Hi, I'm a newbie to the world of HMMs and HMMs in R. I've had a look at the hmm package and the RHmm package but I couldn't see anything straightforward on how a labelled sequential dataset with observed values and underlying states might be used to construct and train a HMM based on that data and no pre-computed values for the transition, emission or initial state distributions. Does

Hello, I know that Mclust does the fitting on its own but I am trying to implement an optimization with the aim to generate a the mixture gaussian with the combine moments as closed as possible to the moment of my return distribution. The objective is to Min Abs((Mean Ret - MeanFit)/Mean Fit) + Abs((Std Ret -Stdev Fit)/Stdev) + Abs((Sk Ret-Sk fit)/Sk Fit) + Abs((Kurt Ret- Kurt Fit)) Taking

Hello, I have initial parameters for HMM model and I want to use depmixS4 package. The parameters are in the form intial_prob_matrix=matrix(c(0.07614213, 0.45177665, 0.47208122), nrow=1, ncol=3, byrow = TRUE) transition_matrix=matrix(c(0.46666667,0.46666667,0.06666667, 0.06741573,0.5617978,0.37078652, 0.02173913,0.3478261,0.63043478), nrow = 3, ncol =

Hello, all-- I am having some fun playing with the graphing in quantmod-- very nice! I am writing a function to calculate (and hopefully plot) support and resistance lines, but the usual plot call of "abline(h=value)" does not seem to work. Here's my code: require(quantmod) AAPL<-getYahooData("AAPL") candleChart(AAPL,subset="last 3

Dear R users, I would like to know wether there exists a package or routine within the R environment which allows to test for multiple Gaussians in a given distribution: I have a histogram of some variable which seems to be tri - or at least bimodal, so I would like to test if this is significant to some level and/or estimate the underlying possible distributions. Thank you very much, Yours

Hi, I need to draw hmm's in R.I also need to include a small plot near each state of the hmm.How can I do this? -- Rajesh.J [[alternative HTML version deleted]]

Hello, all... I've been playing with the TTR package and quantmod, and I'm loading the Chicago Board of Exchange put/call ratio data via a simple read.csv call... CBOEtotal<-read.csv(file=" http://www.cboe.com/publish/ScheduledTask/MktData/datahouse/totalpc.csv ",skip=1) this gives me a data frame with columns.... > names(CBOEtotal) [1] "Trade_date"

Recently, I have been working with some data that look like two overlapping gaussian distributions. I would like to either 1) determine the mean and SD for each of the two distributions OR 2) get some (bayesian ?) statistic that estimates how likely an observation is to belong to the left-hand or right-hand distribution In case I'm using the wrong language, my data looks something like

spData <- as.vector(getYahooData("SPY", start="19900101",end="20081231")[,"Close"]) > cr <- cReturns(spData, ttr = "sma", params=c(20)) Error in ind[t - k] <- pos[t - k + 1] - pos[t - k] : replacement has length zero I am getting the above error when running the cReturns function. Any ideas on this? How does one drill down into

Hello, I am using the parameterized EM- Algorithm in clustering context. I use generated datasets, containing 5000 points in 20 dimensions with 5 clusters. The following poses a problem for me: emEst<-me(modelName="EEV",data=data[,1:numDimensions],z=unmap(x)) emEst<-me(modelName="VEV",data=data[,1:numDimensions],z=unmap(x)) coordProj(main="Model

Hey, Dose R include some package for Gaussian Mixture Model data generation and parameters estimation? Now I want to assign lots of multivariate data into a GMM model. So just wondering if given sample data, can we use some functions to estimate the compoents' density function. Thanks for your support. Fred -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-

Hey, I am confused about how to generate the sample data from a mixture of Multivariate Gaussian ditribution. For example, there are 2 component Gaussian with prior probability of 0.4 and 0.6, the means and variances are u1=[1 1]', Cov1=[1 0;0 1] and u2=[-1 -1]', Cov2=[1 0;0 1] repectively. So how can I generate a sample of 500 data from the above mixture distribution? Thanks. Fred

Dear R users: I am wondering how to ask for *fixed* number of distributions under parameterized Gaussian mixture model. I know that em() and some related functions can predict the parameterized Gaussian mixture model. However, there seems no parameter to decide number of distributions to be mixed (if we known the value in advance). That is, assume I know the (mixed)data is from 3 different