texreg (version 1.31)

extract: Extract coefficients and GOF measures from a statistical object

Description

Extract coefficients and GOF measures from a statistical object.

Usage

extract(model, ...)
extract.aftreg(model, include.loglik = TRUE, include.lr = TRUE, 
    include.nobs = TRUE, include.events = TRUE, include.trisk = TRUE, 
    ...)
extract.betareg(model, include.precision = TRUE, 
    include.pseudors = TRUE, include.loglik = TRUE, 
    include.nobs = TRUE, ...)
extract.brglm(model, include.aic = TRUE, include.bic = TRUE, 
    include.loglik = TRUE, include.deviance = TRUE, 
    include.nobs = TRUE, ...)
extract.btergm(model, conf.level = 0.95, ...)
extract.clm(model, include.thresholds = TRUE, include.aic = TRUE, 
    include.bic=TRUE, include.loglik = TRUE, include.nobs = TRUE, 
    ...)
extract.clogit(model, include.aic = TRUE, include.rsquared = TRUE, 
    include.maxrs = TRUE, include.events = TRUE, 
    include.nobs = TRUE, include.missings = TRUE, ...)
extract.coeftest(model, ...)
extract.coxph(model, include.aic = TRUE, include.rsquared = TRUE, 
    include.maxrs=TRUE, include.events = TRUE, 
    include.nobs = TRUE, include.missings = TRUE, 
    include.zph = TRUE, ...)
extract.coxph.penal(model, include.aic = TRUE, 
    include.rsquared = TRUE, include.maxrs = TRUE, 
    include.events = TRUE, include.nobs = TRUE, 
    include.missings = TRUE, include.zph = TRUE, ...)
extract.dynlm(model, include.rsquared = TRUE, include.adjrs = TRUE, 
    include.nobs = TRUE, include.fstatistic = FALSE, ...)
extract.ergm(model, include.aic = TRUE, include.bic = TRUE, 
    include.loglik = TRUE, ...)
extract.fGARCH(model, include.nobs = TRUE, include.aic = TRUE, 
    include.loglik = TRUE, ...)
extract.gam(model, include.smooth = TRUE, include.aic = TRUE, 
    include.bic = TRUE, include.loglik = TRUE, 
    include.deviance = TRUE, include.dev.expl = TRUE, 
    include.dispersion = TRUE, include.rsquared = TRUE, 
    include.gcv = TRUE, include.nobs = TRUE, 
    include.nsmooth = TRUE, ...)
extract.gamlss(model, robust = FALSE, include.nobs = TRUE, 
    include.nagelkerke = TRUE, include.gaic = TRUE, ...)
extract.gee(model, robust = TRUE, include.dispersion = TRUE, 
    include.nobs = TRUE, ...)
extract.glm(model, include.aic = TRUE, include.bic = TRUE, 
    include.loglik = TRUE, include.deviance = TRUE, 
    include.nobs = TRUE, ...)
extract.glmerMod(model, naive = FALSE, nsim = 1000, 
    conf.level = 0.95, include.aic = TRUE, include.bic = TRUE, 
    include.loglik = TRUE, include.deviance = TRUE, 
    include.nobs = TRUE, include.groups = TRUE, 
    include.variance = TRUE, ...)
extract.gls(model, include.aic = TRUE, include.bic = TRUE, 
    include.loglik = TRUE, include.nobs = TRUE, ...)
extract.gmm(model, include.obj.fcn = TRUE, 
    include.overidentification = FALSE, include.nobs = TRUE, ...)
extract.hurdle(model, beside = FALSE, include.count = TRUE, 
    include.zero = TRUE, include.aic = TRUE, include.loglik = TRUE, 
    include.nobs = TRUE, ...)
extract.ivreg(model, include.rsquared = TRUE, include.adjrs = TRUE, 
    include.nobs = TRUE, include.fstatistic = FALSE, ...)
extract.lm(model, include.rsquared = TRUE, include.adjrs = TRUE, 
    include.nobs = TRUE, include.fstatistic = FALSE, ...)
extract.lme(model, include.aic = TRUE, include.bic = TRUE, 
    include.loglik = TRUE, include.nobs = TRUE, ...)
extract.lme4(model, naive = FALSE, nsim = 1000, 
    conf.level = 0.95, include.aic = TRUE, include.bic = TRUE, 
    include.loglik = TRUE, include.deviance = TRUE, 
    include.nobs = TRUE, include.groups = TRUE, 
    include.variance = TRUE, ...)
extract.lmerMod(model, naive = FALSE, nsim = 1000, 
    conf.level = 0.95, include.aic = TRUE, include.bic = TRUE, 
    include.loglik = TRUE, include.deviance = TRUE, 
    include.nobs = TRUE, include.groups = TRUE, 
    include.variance = TRUE, ...)
extract.lmrob(model, include.nobs = TRUE, ...)
extract.lnam(model, include.rsquared = TRUE, include.adjrs = TRUE, 
    include.aic = TRUE, include.bic = TRUE, include.loglik = TRUE, 
    ...)
extract.lrm(model, include.pseudors = TRUE, include.lr = TRUE, 
    include.nobs = TRUE, ...)
extract.maBina(model, ...)
extract.mer(model, naive = FALSE, nsim = 1000, 
    conf.level = 0.95, include.aic = TRUE, include.bic = TRUE, 
    include.loglik = TRUE, include.deviance = TRUE, 
    include.nobs = TRUE, include.groups = TRUE, 
    include.variance = TRUE, ...)
extract.multinom(model, include.pvalues = TRUE, include.aic = TRUE, 
    include.bic = TRUE, include.loglik = TRUE, 
    include.deviance = TRUE, include.nobs = TRUE, ...)
extract.negbin(model, include.aic = TRUE, include.bic = TRUE, 
    include.loglik = TRUE, include.deviance = TRUE, 
    include.nobs = TRUE, ...)
extract.nlme(model, include.aic = TRUE, include.bic = TRUE, 
    include.loglik = TRUE, include.nobs = TRUE, ...)
extract.nlmerMod(model, naive = FALSE, nsim = 1000, 
    conf.level = 0.95, include.aic = TRUE, include.bic = TRUE, 
    include.loglik = TRUE, include.deviance = TRUE, 
    include.nobs = TRUE, include.groups = TRUE, 
    include.variance = TRUE, ...)
extract.phreg(model, include.loglik = TRUE, include.lr = TRUE, 
    include.nobs = TRUE, include.events = TRUE, include.trisk = TRUE, 
    ...)
extract.plm(model, include.rsquared = TRUE, include.adjrs = TRUE, 
    include.nobs = TRUE, ...)
extract.pmg(model, include.nobs = TRUE, ...)
extract.polr(model, include.thresholds = FALSE, include.aic = TRUE, 
    include.bic = TRUE, include.loglik = TRUE, 
    include.deviance = TRUE, include.nobs = TRUE, ...)
extract.rem.dyad(model, include.nvertices = TRUE, 
    include.events = TRUE, include.aic = TRUE, include.aicc = TRUE, 
    include.bic = TRUE, ...)
extract.rlm(model, include.nobs = TRUE, ...)
extract.rq(model, include.nobs = TRUE, include.percentile = TRUE, 
    ...)
extract.sclm(model, include.thresholds = TRUE, include.aic = TRUE, 
    include.bic = TRUE, include.loglik = TRUE, include.nobs = TRUE, 
    ...)
extract.sienaFit(model, include.iterations = TRUE, ...)
extract.simex(model, jackknife = TRUE, include.nobs = TRUE, ...)
extract.stergm(model, beside = FALSE, include.formation = TRUE, 
    include.dissolution = TRUE, include.nvertices = TRUE, 
    include.aic = FALSE, include.bic = FALSE, 
    include.loglik = FALSE, ...)
extract.survreg(model, include.aic = TRUE, include.bic = TRUE, 
    include.loglik = TRUE, include.deviance = TRUE, 
    include.nobs = TRUE, ...)
extract.survreg.penal(model, include.aic = TRUE, 
    include.bic = TRUE, include.loglik = TRUE, 
    include.deviance = TRUE, include.nobs = TRUE, ...)
extract.svyglm(model, include.aic = FALSE, include.bic = FALSE, 
    include.loglik = FALSE, include.deviance = TRUE, 
    include.dispersion = TRUE, include.nobs = TRUE, ...)
extract.systemfit(model, include.rsquared = TRUE, 
    include.adjrs = TRUE, include.nobs = TRUE, ...)
extract.texreg(model, ...)
extract.tobit(model, include.aic = TRUE, include.bic = TRUE, 
    include.loglik = TRUE, include.deviance = TRUE, 
    include.nobs = FALSE, include.censnobs = TRUE, include.wald=TRUE, 
    ...)
extract.weibreg(model, include.loglik = TRUE, include.lr = TRUE, 
    include.nobs = TRUE, include.events = TRUE, include.trisk = TRUE, 
    ...)
extract.zelig(model, include.aic = TRUE, include.bic = TRUE, 
    include.loglik = TRUE, include.deviance = TRUE, include.nobs = TRUE, 
    include.rsquared = TRUE, include.adjrs = TRUE, include.fstatistic = TRUE, 
    ...)
extract.zeroinfl(model, beside = FALSE, include.count = TRUE, 
    include.zero = TRUE, include.aic = TRUE, include.loglik = TRUE, 
    include.nobs = TRUE, ...)

Arguments

model

A statistical model object.

beside

If available: should the model terms be arranged below each other or beside each other (default)? For example, in a stergm model, the formation and dissolution coefficients can be arranged in two columns of the table.

conf.level

Confidence level (1 - alpha) for computing confidence intervals.

include.adjrs

If available: should the adjusted R-squared be reported?

include.aic

If available: should Akaike's information criterion (AIC) be reported?

include.aicc

If available: should AICc be reported? This is a version of AIC with a correction for finite sample sizes.

include.bic

If available: should the Bayesian information criterion (BIC) be reported?

include.censnobs

If available: should the total, right-censored, left-censored, and uncensored number of observations be reported?

include.count

If available: should the count model of a zero-inflated or hurdle regression be included in the coefficients block (before the zero-inflation or zero hurdle model)?

include.dev.expl

If available: should the deviance explained be reported?

include.deviance

If available: should the deviance be reported?

include.dispersion

If available: should the dispersion or scale parameter be reported?

include.dissolution

If available: should the coefficients for the dissolution phase in a STERGM be reported?

include.events

If available: should the number of events be reported (in survival models)?

include.formation

If available: should the coefficients for the formation phase in a STERGM be reported?

include.fstatistic

If available: should the F statistic be reported?

include.gaic

If available: should the Generalized Akaike's information criterion (GAIC) be reported?

include.gcv

If available: should the GCV score be reported (in GAMs)?

include.groups

If available: should the number of groups be reported?

include.iterations

If available: should the number of iterations be included?

include.loglik

If available: should the log-likelihood be reported?

include.lr

If available: should the likelihood ratio test be reported?

include.maxrs

If available: should the maximum possible R-squared be reported?

include.missings

If available: should the number of missing observations be reported (in survival models)?

include.nagelkerke

If available: should Nagelkerke's R-squared be reported?

include.nobs

If available: should the number of observations be reported?

include.nsmooth

If available: should the number of smooth terms be reported (in GAMs)?

include.nvertices

If available: should the number of vertices be reported in a statistical network model?

include.obj.fcn

If available: should the value of the objective function (= criterion function) be reported (for gmm objects)? More precisely, this returns E(g)var(g)^{-1}E(g).

include.overidentification

If available: should the J-test for overidentification be reported (for gmm objects)?

include.percentile

If available: should the percentile (tau) be reported?

include.precision

If available: should the precision estimates of a betareg fit (the phi coefficients) be reported as part of the coefficients block?

include.pseudors

If available: should the pseudo R-squared be reported?

include.pvalues

If available: should the p values be reported (naive p values are not recommended for lme4 models, but see also the mcmc.pvalues argument)?

include.rsquared

If available: should R-squared be reported?

include.smooth

If available: should the smooth terms of a GAM be reported? If they are reported, the EDF value is reported as the coefficient, and DF is included in parentheses (not standard errors because a chi-square test is used for the smooth terms).

include.thresholds

If available: should the threshold parameters (that is, the intercepts for the class boundaries) be reported in ordinal models?

include.trisk

If available: should the total time at risk be reported (in event-history models)?

include.variance

If available: should group variances be reported?

include.wald

If available: should the Wald statistic be included?

include.zero

If available: should the zero-inflation model of a zero-inflated regression or the zero hurdle model of a hurdle regression be included in the coefficients block (after the count model)?

include.zph

If available: should the Cox proportional hazards assumption be tested (resulting in a p value indicating whether the proportional hazards assumption of the model is violated)?

jackknife

If available: use Jackknife variance instead of Asymptotic variance.

naive

If available: use naive p values and standard errors.

nsim

In linear mixed effects models: the MCMC sample size from which confidence intervals are derived (for old versions of the lme4 package), or the number of simulations on which bootstrapped confidence intervals should be based (for newer versions

robust

If available: report robust instead of naive standard errors.

...

Custom parameters.

Details

extract() is a generic function which extracts coefficients and GOF measures from statistical objects. There are several extract functions for the specific model types, which are called by the generic extract function if it encounters a model known to be handled by the specific function. The output is used by the texreg function.

The various extract functions can also be used directly on a statistical model to convert them into texreg objects.

References

Leifeld, Philip (2013). texreg: Conversion of Statistical Model Output in R to LaTeX and HTML Tables. Journal of Statistical Software, 55(8), 1-24. http://www.jstatsoft.org/v55/i08/.

See Also

texreg-package texreg extract-methods