similar to: Seasonal ARIMA question - stat package (formerly ts)

Dear All, I am trying to fit a structural time series model using the StructTS function (package stats) with only 2 seasons (summer and winter). More than 2 seasons work fine but with 2 seasons I get this error: > fit <- StructTS(y.ts, type="BSM") Error in T[cbind(ind + 1L, ind)] <- 1 : subscript out of bounds I have looked at Prof. Ripley's 2002 RNews article but cannot

Hi, I am interested in using the numeric output from the "gradient" attribute of deriv's output in subsequent analyses. But, I have so far been unable to determine how to do so. I will use the example from the deriv help to illustrate. > ## function with defaulted arguments: > (fx <- deriv(y ~ b0 + b1 * 2^(-x/th), c("b0", "b1", "th"),

Hello, The likelihood includes two parameters to be estimated: lambda (=beta0+beta1*x) and alpha. The algorithm for the estimation is as following: 1) with alpha=0, estimate lambda (estimate beta0 and beta1 via GLM) 2) with lambda, estimate alpha via ML estimation 3) with updataed alpha, replicate 1) and 2) until alpha is converged to a value I coded 1) and 2) (it works), but faced some

Hi, I'm trying to get maximum likelihood estimates of \alpha, \beta_0 and \beta_1, this can be achieved by solving the following three equations: n / \alpha + \sum\limits_{i=1}^{n} ln(\psihat(i)) - \sum\limits_{i=1}^{n} ( ln(x_i + \psihat(i)) ) = 0 \alpha \sum\limits_{i=1}^{n} 1/(psihat(i)) - (\alpha+1) \sum\limits_{i=1}^{n} ( 1 / (x_i + \psihat(i)) ) = 0 \alpha \sum\limits_{i=1}^{n} (

Dear All I'm attempting to speed up my model building procedure, but need some help with the loops I've created...please bear with me through the explanation! My basic model call is something like: m0sulf.nlme<-nlme(conc~beta0*exp(-beta1*day)+beta2*exp(-beta3*day), data=m0sulf, fixed=(beta0+beta1+beta2+beta3~1),

Hi all, I am trying to create simulated data for exploring reclassfication measures in a logistic setting with two continuous predictors and I would like to set the average population probability of outcome rather than the logistic beta0. Is there a way to find a beta0 that will generate the desired average population probability of outcome given x,y and their odds ratios? #Here is an

I use "while" loop but it produces an errro. I have no idea about this. Error in "[<-"(`*tmp*`, i, value = numeric(0)) : nothing to replace with The problem description is The likelihood includes two parameters to be estimated: lambda (=beta0+beta1*x) and alpha. The algorithm for the estimation is as following: 1) with alpha=0, estimate lambda (estimate beta0

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

Background: I administer a cluster of RedHat EWS 3U4 Linux workstations at a university. I built R 2.1.0 from source: ./configure \ --prefix=/sscc/opt/R-2.1.0 \ --with-blas=no \ 2>&1 \ | tee NUInstall.configure R is now configured for i686-pc-linux-gnu Source directory: . Installation directory: /sscc/opt/R-2.1.0 C compiler:

Dear Users, I am new to both of things, so do not blame me too much... I am busy with semiparametric regression and use WinBUGS to sample posteriors. The code to call Winbugs is as follows: data <- list("y","X","n","m") #My variables inits.beta <- rep(0,K) inits.beta0 <- 0 inits <-

Dear list, After running for a while, it crashes and gives the following error message: can anybody suggest how to deal with this? Error in if (ratio0[i] < log(runif(1))) { : missing value where TRUE/FALSE needed ################### original program ######## p2 <- function (Nsim=1000){ x<- c(0.301,0,-0.301,-0.602,-0.903,-1.208, -1.309,-1.807,-2.108,-2.71) # logdose

Hi all, Does anyone know how ts objects ts(base) can be used within the 'seas' package? I would like to obtain seasonal statistics of regular time-series and for example look at the result of the plot.seas.var() function or use the change function() to look at change between periods or time-series. The nottem time-series are similar to the time-series we are analyzing (but with

Dear all, I am trying to evaluate the optimisation behaviour of a function. Originally I have optimised a model with real data and got a set of parameters. Now I am creating simulated data sets based on these estimates. With these simulations I am estimating the parameters again to see how variable the estimation is. To this end I have written a loop which should generate a new simulated data

Respected Sir, With reference to my mail to you and the reply mail by you dated 9th and 16th January, 2013, I am sending the reproducible code in the attached document named " MODIFIED ANS ". I am also attaching the txt file named "hazModel", which is required to save in my documents folder to run the program. The file also contains the error message

Dear all,I am a new user of WinBUGS and need your help. After running the following code, I got parameters of beta0 through beta4 (stats, density), but I don't know how to get the prediction of the last value of h, the variable I set to NA and want to model it using the following code.Does anyone can given me a hint? Any advice would be greatly appreciated.Best

Hi, I am trying to covert a Winbugs code into R code. Here is the winbugs code model{# model’s likelihoodfor (i in 1:n){time[i] ~ dnorm( mu[i], tau ) # stochastic componenent# link and linear predictormu[i] <- beta0 + beta1 * cases[i] + beta2 * distance[i]}# prior distributionstau ~ dgamma( 0.01, 0.01 )beta0 ~ dnorm( 0.0, 1.0E-4)beta1 ~ dnorm( 0.0, 1.0E-4)beta2 ~ dnorm( 0.0, 1.0E-4)#

David's answers were correct. You are looking deep into the code when there is no reason to to so. 1. h(t|(X=x,Z=z)) = exp(Beta0 + XZBeta1) Most statisticians will tell you that this is an unwise model. The reason is that if you replace X with "X+1" the fit changes, which is almost never desirable. What if someone coded your dummy variable as 1/0 instead of 0/1 -- wouldn't

Hi, I am using the sim1000G R package to simulate data for case/control study. I can not figure out how to manipulate this code to be able to generate 10% or 50% causal SNPs in R. This is whole code provided as example on GitHub: library(sim1000G) vcf_file = "region-chr4-357-ANK2.vcf.gz" #nvariants = 442, ss=1000 vcf = readVCF( vcf_file, maxNumberOfVariants = 442 ,min_maf =

Dear all, My code is presented as the following. library(MASS) library(rmutil) library(repeated) library(lme4) library(arm) #install.packages("Zelig", repos = "http://gking.harvard.edu") library(Zelig) rm(list = ls()) beta0<-2.5 beta1<--0.3 sigs2<-0.5 I<-4 #numberpatients<-c(40,100,160,200,400,600) numberpatients<-c(1000) #numberpatients<-3 times<-1

Hello R-Experts, I am facing the problem that I have to estimate several parameters for a lot of different dependent variables. One single regression looks something like this: y = beta0 + beta1 * x1 + beta2 * x2 + beta3 * x1 * x2 + beta4 * x4 + beta5 * lag(x4,-1) where y is the dependent variable and xi are the independent ones. Important to me are the different estimates of betai and their