search for: u_i

Hi Is there any way to fit a bivaraite longitudinal mixed model using R. I have a data set with col names resp1 (Y_ij1), resp2 (Y_ij2), timepts (t_ij), unit(i) j=1,2,..,m and i=1,2,..n. I want to fit the following two models Model 1 Y_ij1, Y_ij2 | U_i = u_i ~ N(alpha + u_i + beta1*t_ij, Sigma) U_i ~ iid N(0, sigu^2) Sigma = bivariate AR structure alpha and beta are vectors of order 2. Model 2 Y_ij, Y_ij2 | U_i = u_i ~ N(alpha + u_i + beta1*t_ij + U_i2*t_ij, Sigma) U_i=(U_i1,U_i2) ~ iid N(0, Omega) Sigma =bivariate AR structure Here alpha...

...to linearly regress y on x. There does not seem to be an option to 'lm' which can deal with such data directly, though the regression problem could be algebraically expressed in these terms. One way of fudging it would be to replace each m_i by a set of n_i numbers Y_i constructed as u_i <- rnorm(ni) Y_i <- m_i + s_i*(u_i - mean(u_i))/sd(u_i) and associate these with X_i <- rep(x_i,n_i), thereby constructing a regression-equivalent set of pseudo "raw data" which could be fed to lm(Y~X). However, this strikes me as cumbersome, at least, and even ugly! Is ther...

...Number of groups = 545 R-sq: within = 0.1799 Obs per group: min = 8 between = 0.1860 avg = 8.0 overall = 0.1830 max = 8 Random effects u_i ~ Gaussian Wald chi2(14) = 957.77 corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000 ------------------------------------------------------------------------------ lwage | Coef. Std. Err. z P>|z| [95% Conf. Inter...

...lternative is to write the likelihood down and use optim or a similar function to get the MLE of the parameters of the model. I was also trying to compute Small Area Estimators using area level models, so that I have one observation per area. I tried to fit the following model: Yhat_i = X beta + u_i + e_i where Yhat is a direct estimator of the target variable, X, the area- level covariates, u_i the random effects independent and distributed as N(0, sigma2_u) and e_i the 'error' terms, which are distributed as N(0, sigma2_e/n_i), where n_i is the sample size in area i. This model s...

...matrix of nrow=30 and ncoll=54,and I would like to generate 50 simulations with tha same size of the matrix!!!That is to say that I want to generate 50 matrices -for my 50 simulations - with the same dimensions! I took my 1st matrix according to the formula that I want to implement: D<-mean_m + U_i*mat_DELTA mean_m:vector U_I:vector mat_Delta:matrix(54,30) Here is the loop: #Define Simulations(IS) #Define Time Step(IT) vec_IS<-c(1:50) vec_IT<-c(1:30) mat_delta<-matrix(nrow=54, ncol=30) mat_DELTA<-matrix(nrow=54, ncol=30) mat_Yr_m<-matrix(nrow=54, ncol=30) D<-matrix(nrow=3...

...can be put out into a SAS data set. How do I do the same with lme-objects? For example, I can see the estimated variance matrix when typing the name of the lme-object, but that does not give me a "handle" on the matrix. 2) I am trying to fit a mixed model, say y_{ij} = X_{ij}\beta + U_i + e_{ij} where Y_{ij}, U_i and e_{ij} all are D-dimensional, and U_i ~ N_D(0,G), and e_{ij} ~ N(0,R). This is easy to do in proc mixed by including a factor keeping track of how the elements of y_{ij} are "stacked" on top of each other. This factor is then used in the model, the random a...

...Obs per group: min = 1 between = 0.9122 avg = 3.0 overall = 0.9635 max = 3 F(11,97301) = 949024.29 corr(u_i, Xb) = 0.2166 Prob > F = 0.0000 ------------------------------------------------------------------------------ lnLP | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+----------------------------------------------...

In a classical meta analysis model y_i = X_i * beta_i + e_i, data {y_i} are assumed to be independent effect sizes. However, I'm encountering the following two scenarios: (1) Each source has multiple effect sizes, thus {y_i} are not fully independent with each other. (2) Each source has multiple effect sizes, each of the effect size from a source can be categorized as one of a factor levels

...o do. Suppose we are given a stream of pseudo-random numbers, U_1, U_2... intended to simulate a sequence of iid rvs from Unif[0,1], and a probability distribution (specified as a pdf f or cdf F), and are required to obtain a sequence of rvs X_1, X_2... such that X_i is a deterministic function of U_i. Let us further suppose that F^{-1} is difficult to calculate, so one does not simply do X_i = F^{-1}(U_i). Now, a possible alternative would be use U_i to set a random seed for a sequence of pseudo-random numbers, V_{i1}, V_{i2},... Then we can use this V_{ij} stream for every i to simulate F us...

Dear R users, I am using the function "constrOptim" to minimize the -1*log-likelihood where \beta_i>=0 i=1,...,p and \beta_0 is unconstrained. I construct u_i as 0 0 0 ... 0 0 1 0 ... 0 0 0 1 ... 0 . . . ... 0 . . . ... 0 . . . ... 0 0 0 0...

...Obs per group: min = 1 between = 0.9122 avg = 3.0 overall = 0.9635 max = 3 F(11,97301) = 949024.29 corr(u_i, Xb) = 0.2166 Prob > F = 0.0000 ------------------------------------------------------------------------------ lnLP | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+-----------------------------------------...

...nction R at MLE by declaring hessian=TRUE in the calls to the function optim. I would like to ask if it is advisable to get an approximate hessian for the funcrion f as follows: f''(theta)=R''(theta|theta_k)-B''(theta) where B''(theta)=mu*sum((g(theta_k)/g(theta)^2)u_i*u_i^T) denotes the second derivative of the barrier function (following the notation given by Lange (1999)) where R''(theta|theta_k) will be replaced by the approximate hessian at MLE returned by optim. Thanks in advance, Spyridoula Tsonaka Doctoral Student Biostatistical Centre Katholi...

The compositional data xi=(x_i1,x_i2,...,x_in), for each fixed i , xij>0, and sum(xij)=1; I want to compare the mean( u_i) of several groups i.e. H0: u_1=u_2=...=u_N or H0: u_11=u_21=...=u_N1 Are there any ANOVA tpye tools to do this work in R? Thanks, WEN S Q [[alternative HTML version deleted]]

.... failure times subject to right censoring) that is observed for N subjects: Y_i, I = 1, 2, ..., N. Note that Y_i = min(T_i, C_i), where T_i is the true failure time and C_i is the censored time. Let us assume that C_i is independent of T_i. Now, I would like to generate another random variable U_i, I = 1, 2, ..., N, which is correlated with T. In other words, I would like to generate U from the conditional distribution [U | T=t]. One approach might be to assume that the joint distn [T, U] is bivariate lognormal. So, the marginals [T] and [U], as well as the conditional [U | T] are also lo...

...122 avg = 3.0 > > > > overall = 0.9635 max = 3 > > > > > > > > F(11,97301) = 949024.29 > > > > corr(u_i, Xb) = 0.2166 Prob > F = 0.0000 > > > > > > > > ------------------------------------------------------------------------------ > > > > lnLP | Coef. Std. Err. t P>|t| [95% Conf. Interval] > &...

...on only a random draw, not all of doctors' patients. I am interested in the, say, median number of patients of doctors. Does it suffice to use the "nid" option in summary.rq? More specifically, if the model generating the number of patients, say, r_i, of doctor i is r_i = const + u_i, then I think I would obtain the median of the number of doctors' patients using rq(r~1, ...) and plugging this into summary.rq() using the option se="iid". Unfortunately, I don't observe r_i in the data but, instead, in the data I only have a fraction p of these r_i patients. In...

Dear useRs (called Frank Harrell, most likely), after having preached for years to my medical colleagues to be cautious with stepwise selection procedures, they chanted back asking for an alternative when using mixed models. There is a half dozen laXXX packages around for all types of linear models, but as far I see there is none for mixed models such as lme. Even boot.stepAIC (which I

...f0 = p*(1-s)^4 + (1-p)*(1-t)^4; f1 = 4*p*s*(1-s)^3 + 4*(1-p)*t*(1-t)^3; f2 = 6*p*s^2*(1-s)^2 + 6*(1-p)*t^2*(1-t)^2; f3 = 4*p*s^3*(1-s) + 4*(1-p)*t^3*(1-t); f4 = p*s^4 + (1-p)*t^4; The polynomial f_i represents the probability of seeing i successes. Suppose we repeat this experiment 1000 times, and u_i is the number of times we saw i successes. The likelihood of this event is f_0^u_0*f_1^u_1*f_2^u_2*f_3^u_3*f_4^u_4, and we seek to find those parameter values for s,t,p which maximize the likelihood. My Singular package has as input the 5 polynomials and a data vector u. For the particular exampl...

The compositional data xi=(x_i1, x_i2,..., x_in), for each fixed i , xij>0, and sum(xij)=1; I want to compare the mean( u_i) of several groups i.e. H0: u_1=u_2=...=u_N or Hj0: u_1j=u_2j=...=u_Nj Are there any ANOVA tpye tools to do this work in R? Thanks, WEN S Q [[alternative HTML version deleted]]

...I'm hoping someone can help shed some light on this. I simply want to get asterisk/sangoma to send the same signalling that Paetec does. I am contacting them for details on the setup they are using also. http://www.dialogic.com/manuals/docs/isdn_api_v5.pdf The Dialogic system is using the U_IES[1] with cc_GetCallInfo() to get the information sent from Paetec, our provider, but the information isn't being sent the same way from the Sangoma card, even though I have it set to pri_net in zapata.conf. Alternatively, we can check the UUI[2] message to see if the Dialogic card will recei...