similar to: Error in solve.default(Hessian) : system is computationally singular

Hello R-list-members, I'm trying to model ARMA(0,2) & GARCH(1,1) process using the code below, but according to my textbook, the estimated parameters are wrong. The MA-parameters should be negative. (I've got the same problem using garchFit()). Can anyone tell me what I'm doing wrong? And how can I calculate the hessian matrix? Many thanks, Desislava Kavrakova Code:

Hello, i want to estimate a complex GARCH-model (see below). http://r.789695.n4.nabble.com/file/n4112396/GJR_Garch.png W stands for the Day of the Week Dummies. r stands for returns of stock market indices. I stands for the GJR-term. I need some help with three problems: 1.) implementation of the GJR-term in the variance equation 2.) compute robust covariance matrix

My question is about the "Signif. codes" , the output when I run matcoef =cbind(fit$par, se.coef,tval,2*(1-pnorm(abs(tval)))) dimnames(matcoef)=list(names(tval),c("Estimate","Std.Error","t value","pr(>|t|)")) cat("\nCoefficient(s):\n") printCoefmat(matcoef, digits=4, signif.stars = TRUE) Coefficient(s): Estimate

Dear All, I am running a simulation to obtain coverage probability of Wald type confidence intervals for my parameter d in a function of two parameters (mu,d). I am optimizing it using "optim" method "L-BFGS-B" to obtain MLE. As, I want to invert the Hessian matrix to get Standard errors of the two parameter estimates. However, my Hessian matrix at times becomes

Hi, it would be nice if the VaR package could get expanded to work with more models. I wrote a function which calculates the VaR for a Garch(1,1) AR(2) model. Since I am fairly new to R I am sure it can be written much neater. If someone feels like having a go, here is what I have written library(fSeries) library(tseries) VaR.Garch <- function(data,stockId=1,p=0.01,dt=1) { r <-

Hi, I am trying to write the Realized GARCH model with order (1,1) The model can be describe bellow: r_t = sqrt( h_t) * z_t logh_t = w + b*logh_(t-1) + r*logx_(t-1) logx_t = c + q*logh_t + t1*z_t +t2*(z_t ^2 -1) + u_t and z follow N(0,1) , u follow N(0, sigma.u^2) But I'm troubled with the simulation check for my code. After I simulate data from the model and estimate the data, I

I am using nlm to maximize a likelihood function. When I call the likelihood function (garchLLH) via nlm however, nlm returns the wrong value of the function. When I test the likelihood function manually I get the correct answer. I'm probably doing something really stupid, maybe someone can point it out for me. ###############this is the function i am trying to minimize ############

Dear R users, I need to estimate an ARCH(1,0) model with t-residuals. To do this I use garchFit function from fGarch library. However, I get the following error message: Error in.garchInitParameters (formula.mean = formula.mean, formula.var = formula.var, ): object "alpha" not found I tried to estimate this model with different series, but I always get this error message. For example,

Hello, I want to have a title that will look something like: "Results for \theta=2.1", given that I have a variable theta=2.1, and \theta should show on the screen like the greek letter. I've tried a lot of things: theta <- 2.1 plot(1:10, main=expression(paste("Results for", theta, "=", eval(theta)))) or using bquote plot(1:10, main=paste("Results for

Hi all-- We are planning an intervention study for adolescent alcohol use, and I am planning to use simulations based on a hurdle model (using the hurdle() function in package pscl) for sample size estimation. The simulation code and power code are below -- note that at the moment the "power" code is just returning the coefficients, as something isn't working quite right. The

I'm trying to build R 2.10.1 on a Sun Blade 1000 running Solaris 10 (03/05 release). I've installed iconv 1.13.1 and used: CPPFLAGS="-I /export/home/drkirkby/sage-4.3.3.alpha0/local/include" (which is where iconv is) LDFLAGS= -R/export/home/drkirkby/sage-4.3.3.alpha0/local/lib -L/export/home/drkirkby/sage-4.3.3.alpha0/local/lib The build of R fails as below. gcc

Full_Name: Juan Gea Version: R version 2.6.2 OS: Fedora Core 6 Submission from: (NULL) (79.153.48.49) Abort: objeS <- matrix("AAA",1000000) class(objeS) outTxt <- textConnection("vaClob", open = "w", local = FALSE) dput(objeS,outTxt) close(outTxt) R version 2.6.2 (2008-02-08) Copyright (C) 2008 The R Foundation for Statistical Computing ISBN

If you found my subject heading to be confusing then I'm sure you'll enjoy the example I've included below. I find the apply type functions to be wonderful for avoiding loops but when I use them with existing functions, I end up using loops anyway to extract the vectors I want. I would appreciate it if someone could show me how to avoid these loops. Thanks. EXAMPLE:

Curious to know if this warning is expected behavior, and if so, what is the recommended way instead: > dput(letters, file = stderr()) c("a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p",

Hi All: I had a basic question to ask. I am running R2WinBUGS so that I could automate the running of my model using 1000 simulated datasets. Below is the code I am using. The only problem I am having is the bugs output that comes out shows my parameters as nos with 1 decimal place after. I would want to have the parameters with 5 places after decimal. How would I specify that in my code for

Hi I am trying to use the deriv and eval functions to obtain the value of a function , say "xi-(alpha0+alpha1*gi)" , differentiated with respect to alpha0 and alpha1, in the following way # for gi = 0 > dU1dtheta <- deriv(~ xi-(alpha0+alpha1*gi), c("alpha0","alpha1")) > eval(dU1dtheta) (Intercept) -0.2547153 attr(,"gradient")

Hello, I was using garchFit {fGarch} to fit some GARCH processes. I noticed that the result contains "Log Likelihood" value (right above "Description"), but when I use .. at fit$llh to retrieve Log Likelihood value, the sign switched. I am confused about which value I should choose to report... Any help here? Thanks a lot! Ted -- View this message in context:

* Ralf Hildebrandt <Ralf.Hildebrandt at charite.de>: > > It seems to loop in sha1_loop & hash_format_loop > > The problem occurs in both 2.3 and 2.2 (I just updated to 2.3 to check). I'm seeing the first occurence of that problem on the 10th of october! I was using (prior to the 10th) : 2.3.0~alpha0-1~auto+371 On the 10th I upgraded (16:04) to:

Hello all, I'm trying to create a heatmap using 2 matrices I have: z and v. Both matrices represent different correlations for the same independent variables. The problem I have is that I wish to have the values from matrix z to be represented by color intensity while having the values from matrix v to be represented by circle size. I currently have the following in front of me and an unsure

Dear all, Can any one tell me how can i perform Orthogonal Polynomial Regression parameter estimation in R? -------------------------------------------- Here is an "Orthogonal Polynomial" Regression problem collected from Draper, Smith(1981), page 269. Note that only value of alpha0 (intercept term) and signs of each estimate match with the result obtained from coef(orth.fit). What