search for: hccm

I am trying to calculate robust standard errors for a logit model. I installed the package "car" and tried using hccm.default, but that required an lm object. Is there some way to do a similar operation for a glm object? x <- hccm.default(glm(winner ~ racebl + racehi + raceas + inchi + incmed + edhs + edcol + edba + agec1 + agec4 + sex + margin + regla + regbay + regsc + libcon+ pdem + poth, data = zol,...

Using hccm() I got a heteroscedasticity correction factor on the diagonal of the return matrix, but I don't know how to incorporate this into my linear model: METHOD 1: > OLS1 <- lm(formula=uer92~uer+low2+mlo+spec+degree+hit) Coefficients: Estimate Std. Error t value Pr(>|t|) (In...

dear R experts: I wonder whether it is possible to extend the summary method for the lm function, so that it uses an option "hccm" (well, model "hc0"). In my line of work, it is pretty much required in reporting of almost all linear regressions these days, which means that it would be very nice not to have to manually library car, then sqrt the diagonal, and recompute T-stats; instead, I would love to get ever...

Dear all, Sorry if this is too obvious. I am trying to fit my multiple regression model using lm() Before starting model simplification using step() I checked whether the model presented heteroscedasticity with ncv.test() from the CAR package. It presents it. I want to correct for it, I used hccm() from the CAR package as well and got the Heteroscedasticity-Corrected Covariance Matrix. I am not sure what am I supposed to do with the matrix. I guess I should run my model again telling it to use that matrix but I don't really find the parameter in lm() to tell R so. I guess it should be...

Hi! I am trying to estimate Engel curves using a big sample (>42,000) using lm and taking heteroskedasticity into account by using the summaryHCCM posted here by John Fox (Mon Dec 25 16:01:59 CET 2006). Having used the SIC (with MASS stepAIC) to determine how many powers to use I estimate the model: > # ========================================= > summary.lm(fit.lm.5) Call: lm(formula = energyshare ~ 1 + I(x.log) + I(x.log^2) + I(x.lo...

I would like to know if there is a function that will run a linear regression and report the White (heteroscedasticity consistent) std. errors. I've found the hccm() function in the car library, but that just gives me the White covariance matrix. I'd like to be able to see the White std. errors without having to do much more work, if possible. Thanks, Brian

...h Huberized variance matrix. > I frequently have to run regressions w/ robust variances. In Stata, one > simply adds the word "robust" to the end of the command or > "cluster(cluster.variable)" for a cluster-robust error. In R, there are two > functions, robcov and hccm. I had to run tests to figure out what the > relationship is between them and between them and Stata (robcov w/o cluster > gives hccm's hc0; hccm's hc1 is equivalent to Stata's 'robust' w/o cluster; > etc.). This is rather clearly document on the respective man pages....

Does anybody have any suggestions regarding applying standard regression packages lm(), hccm(), and others within a parallel environment? Most of the packages I've found only deal with iterative processes (bootstrap) or simple linear algebra. While the latter might help, I'd rather not program the estimation code. I'm currently using a IA-64 Teragrid system through UC San Di...

...summary.lme (extended lm) method that [a] adds normalized coefficients and [b] white heteroskedasticity adjusted se's and T's. I believe I already know how to do the programming to do these two, at least in simple unweighted cases. Now my challenges are just [1] to trap weird cases (e.g., hccm dies because the standard errors cannot be computed), and [2] to follow "proper R rules and regulations." I started my experiments by copying summary.lm() to summary.lme. But there is some magic that I do not understand. Apparently, the class(ans) <- "summary.lm" signals...

Hi, I'm using maxlik with functions specified (L, his gradient & hessian). Now I would like determine some robust standard errors of my estimators. So I 'm try to use vcovHC, or hccm or robcov for example but in use one of them with my result of maxlik, I've a the following error message : Erreur dans terms.default(object) : no terms component Is there some attributes to give to maxlik objet for "fitting" the call of vcovHC? Thank you [[alterna...

...; When replying, please edit your Subject line so it is more specific > than "Re: Contents of R-help digest..." > > > Today's Topics: > > 1. Re: how to 'get' an object that is part of a list > (Gabor Grothendieck) > 2. Re: extend summary.lm for hccm? (Achim Zeileis) > 3. Re: extend summary.lm for hccm? (John Fox) > 4. Hmisc - some latex problems (steve) > 5. Re: Hmisc - some latex problems (Frank E Harrell Jr) > 6. Problem to generate training data set and test data set > (Aimin Yan) > 7. No fonts in graphics u...

I am trying to compute the white heteroskedasticity-robust standard errors (also called the Huber standard errors) in a linear model, but I can't seem to find a function to do it. I know that the design library in S+ has something like this (robcov?), but I have not yet seen this library ported to R. Anyone know if there is already a function built into R to do this relatively simple job?

...t "new" package: sandwich 0.1-3. sandwich provides heteroskedasticity (and autocorrelation) consistent covariance matrix estimators (also called HC and HAC estimators). The former are implemented in the function vcovHC() (which was available in strucchange before - and independently in hccm() in John Fox's car package). And the latter are implemented in the function vcovHAC(). This implements sandwich-type estimators in a rather flexible way, allowing for user-defined weights or weight functions. It builds on some of the functionality which was before available in Thomas Lumley...

...ing, Variable recoding, subseting, sorting, merging, transposing, opening data (text and foreign), and saving data Analysis: Frequencies, Descriptives, Contingency tables (and related statistics), one-sample, two-sample, k-sample tests, as well as correlations Models: Linear Models (with optional HCCM), Logistic regression, Generalized Linear Models Since it?s initial release in August, there have been significant changes to the back-end as well as the programmatic interface. This has resulted in increased stability, and made for easier incorporation of Deducer?s R functions into non-GUI progra...

...t "new" package: sandwich 0.1-3. sandwich provides heteroskedasticity (and autocorrelation) consistent covariance matrix estimators (also called HC and HAC estimators). The former are implemented in the function vcovHC() (which was available in strucchange before - and independently in hccm() in John Fox's car package). And the latter are implemented in the function vcovHAC(). This implements sandwich-type estimators in a rather flexible way, allowing for user-defined weights or weight functions. It builds on some of the functionality which was before available in Thomas Lumley...

...ing, Variable recoding, subseting, sorting, merging, transposing, opening data (text and foreign), and saving data Analysis: Frequencies, Descriptives, Contingency tables (and related statistics), one-sample, two-sample, k-sample tests, as well as correlations Models: Linear Models (with optional HCCM), Logistic regression, Generalized Linear Models Since it?s initial release in August, there have been significant changes to the back-end as well as the programmatic interface. This has resulted in increased stability, and made for easier incorporation of Deducer?s R functions into non-GUI progra...

Hi! I need to run ols regressions with Huber-White sandwich estimators and the correponding standard errors, without an intercept. What I'm trying to do is create an ols object and then use the robcov() function, on the order of: f <- ols(depvar ~ ind1 + ind2, x=TRUE) robcov(f) However, when I go f <- ols(depvar ~ ind1 + ind2 -1, x=TRUE) I get the following error: Error in

...is.lm function (mod, hypothesis.matrix, rhs = 0, white.adjust = F, error.SS, error.df) { sumry <- summary(mod, corr = FALSE) s2 <- if (missing(error.SS)) sumry$sigma^2 else error.SS/error.df V <- if (white.adjust == FALSE) sumry$cov.unscaled else hccm(mod, type = white.adjust)/s2 b <- coefficients(mod) L <- if (is.null(dim(hypothesis.matrix))) t(hypothesis.matrix) else hypothesis.matrix q <- nrow(L) SSH <- t(L %*% b - rhs) %*% inv(L %*% V %*% t(L)) %*% (L %*% b - rhs) F <- SSH/(q * s2)...

Readers of this list might be interested in the following commenta about R. In a recent report, by Michael N. Mitchell http://www.ats.ucla.edu/stat/technicalreports/ says about R: "Perhaps the most notable exception to this discussion is R, a language for statistical computing and graphics. R is free to download under the terms of the GNU General Public License (see http://www.r-project.