similar to: assessing goodness of variance prediction

Dear R-users, I am trying to estimate a state space model of the form (1) b_t = G * b_t-1 + w_t w_t ~ N(0,W) (2) y_t= A' * x_t + H' * b_t + v_t v_t ~ N(0,V) (Hamilton 1984: 372) In particular my estimation in state space form looks like (3) a3_t = 1 * a3_t-1 + w_t w_t ~ N(0,W) (4) g_t = (a1, a2) * (1, P_t)' + u_t * a3_t + v_t v_t ~ N(0,V) where g_t is the

Dear r-users! I have another question regarding the dlm package and I would be very happy if someone could give me a hint! I am using the dlm package to get estimates for an endogenous rate of capacity utilization over time. The general form of a state space model is (1) b_t = G * b_t-1 + w_t w_t ~ N(0,W) (2) y_t= A' * x_t + H' * b_t + v_t v_t ~ N(0,V) (Hamilton 1984: 372) The

I am trying to make the following Kalman filter equations work and therefore produce their graphs. v_t=y_t - a_t a_t+1=a_t+K_t*v_t F_t=P_t+sigma.squared.epsilon P_t+1=P_t*(1-K_t)+sigma.squared.eta K_t=P_t/F_t Given: a_1=0,P_1=10^7,sigma.squared.epsilon=15099, sigma.squared.eta=1469.1 I have attached my code,which of course doesnt work.It produces NAs for the Fs,Ks and the a. Can somebody tell me

Hi all, I'm working on a term structure estimation using state-space modeling for 1, 2 and 3 factor models. When I started to read the functions on R, I got to the function ss on the library sspir. From what I understood this function is similar to SsfFit from S-PLUS. But for my models purpose there is something left to be desired. Its formulation follow these equations: *Y_t = F_t^T *

Dear all, I am currently trying to use the "dlm" package for Kalman filtering. My model is very simple: Y_t = F'_t Theta_t + v_t Theta_t = G_t Theta_t-1 + w_t v_t ~ N(0,V_t) = N(0,V) w_t ~ N(0,W_t) = N(0,W) Y_ t is a univariate time series (1x1) F_t is a vector of factor returns (Kx1) Theta_t is the state vector (Kx1) G_t is the identity matrix My first

  Hello, fellow R users,   I recently need to estimate a dynamic linear model in the following form:   For the measurement equation:   Y_t = F_t * a_t + v_t   where Y_t is the observation. It is a 1 by q row vector for each t. F_t is my forecasting variable. It is a 1 by p row vector. a_t is my state variable. It is a p by q MATRIX of parameters with each column of the matrix being regression

Hi, I am trying to figure how to use dlm with constant terms (possibly time-dependent) added to both equations y_t = c_t + F_t\theta_t + v_t \theta_t = d_t + G_t\theta_{t-1} + w_t, in the way that S-PLUS Finmetrics does? Is there any straightforward way to transform the above to the default setup? Thanks, Tsvetan -------------------------------------------------------- NOTICE: If received in

Dear R users, suppose we have a random walk such as: v_t+1 = v_t + e_t+1 where e_t is a normal IID noise pocess with mean = m and standard deviation = sd and v_t is the fundamental value of a stock. Now suppose I want a trading strategy to be: x_t+1 = c(v_t – p_t) where c is a costant. I know, from the paper where this equations come from (Farmer and Joshi, The price dynamics of common

Hi all, Suppose I am fitting an arma(p,q) model to a time series y_t. So, my model should contain (q+1) white noise variables. As far as I know, each of them should have the same variance. How do I get the estimate of this variance by running the arma(y) function (or is there any other way)? Appreciate your help. Thanks, Preetam -- Preetam Pal (+91)-9432212774 M-Stat 2nd Year,

For 2 or 3 stage least squares, what do you R folks do? Follow-up question. My student wants to estimate this. 2 variables are governed by a system of difference equations. His theory is like so. Y_t and X_t are state variables, we want estimates for a, g, b, and h. X_(t+1) = 1 + a X_t + (a/K)* (X_t)^2 - g Y_t X_t Y_(t+1) = b Y_t + h* X_t * Y_t K is perhaps something to estimate, but it

Hi, I've been searching CRAN and the web for a recursive multivariate filter with time-varying coefficients. What I mean is the following: I have a series of square matrices A_t an initial value vector y_0 and I need to compute y_t =A_t%*%y_t-1 As these y_t may diverge quickly and/or lead to underflow problems, the y_t need to be scaled by eg y_t =y_t/sum(y_t-1) Is anyone aware

I have a question and it's only relation to R is that I probably need R after I understand what to do. Both models are delta y_t = Beta + epslion and suppose I have a null hypothesis and alternative hypothesis H_0 : delta y_t = zero + epsilon epsilon is normal ( 0, sigmazero^2 ) H_1 delta y_t = beta + epsilon epsilon is normal ( sigmabeta^2 )

Dear R-help-archive.. I am trying to figure out how to make arima prediction when I have a process involving multivariate time series input, and one output time series (output is to be predicted) .. (thus strictly speaking its an ARMAX process). I know that the arima function of R was not designed to handle multivariate analysis (there is dse but it doesnt handle arma multivariate analysis, only

Dear R users, I wish to manually demean a panel over time and entities. I tried to code the Wansbeek and Kapteyn (1989) transformation (from Baltagi's book Ch. 9). As a benchmark I use both the pmodel.response() and model.matrix() functions in package plm and the results from using dummy variables. As far as I understood the transformation (Ch.3), Q%*%y (with y being the dependent variable)

This is not an R question but if anyone can help me, it's much appreciated. Suppose I have a series ( stationary ) y_t and a series x_t ( stationary )and x_t has variance sigma^2_x and epsilon is normal (0, sigma^2_epsilon ) and the two series have the relation y_t = Beta*x_t + epsilon My question is if there are particular values that sigma^2_x and sigma^2_epsilon have to take in

Dear R-Users! I am using the svm function (e1071 package) to classify two groups using a set of 180 indicator variables. Now I am confused about the cross-validation procedure. (A) On one hand I use the setting cross=10 in the svm function to run 10 cross-validation iterations and to get an estimate of the svm's performance in prediction. (B) On the other hand most tutorials I found

Is there a function in r that let's you efficiently invert a positive definite symmetric Block Toeplitz matrix? My matrices are the covariance matrices of observations of a multivariate time series and can be 1000*1000 or larger. I know the package 'ltsa' which seems to use the Trench algorithm to compute the inverse of a Toeplitz matrix. I am looking for a so to say

hi: you should probably send below to R-Sig-Finance because there are some econometrics people over there who could also possibly give you a good answer and may not see this email ? Also, there's package called mar ( I think that's the name ) that may do what you want ? Finally, I don't know how to do it but I think there are ways of converting a multivariate arima into the

This is a statistics question not an R question. When calculating the sample correlation coefficient cor(x_t,y_t) between say two variables, x_t and y_t t=1,.....n ( one can assume that the variables are in time but I don't think this really matters for the question ), does someone know where I can find any piece of literature that says that each (x_j,y_j) pair has To be independent from the

I am having a problem with the package SSPIR. The code below illustrates it. I keep getting the message: "Error in y - f : non-conformable arrays." I tried to tweak the code below in many different ways, for example, substituting rbind for cbind, and sometimes I get a different error message, but I could not find a variation of this code that would work. Any help will be greatly