\name{unmarkedFit-class} \Rdversion{1.1} \docType{class} \alias{unmarkedFit} \alias{unmarkedFit-class} \alias{getData} \alias{getData,unmarkedFit-method} \alias{hessian} \alias{hessian,unmarkedFit-method} \alias{hist,unmarkedFitDS-method} \alias{mle} \alias{mle,unmarkedFit-method} \alias{names,unmarkedFit-method} \alias{nllFun} \alias{nllFun,unmarkedFit-method} \alias{parboot,unmarkedFit-method} \alias{parboot,unmarkedFitOccuMulti-method} \alias{plot,unmarkedFit,missing-method} \alias{plot,unmarkedFitOccuMulti,missing-method} \alias{plot,unmarkedFitGDR,missing-method} \alias{plot,unmarkedFitIDS,missing-method} \alias{plot,unmarkedFitOccuCOP,missing-method} \alias{profile,unmarkedFit-method} \alias{residuals,unmarkedFit-method} \alias{residuals,unmarkedFitOccu-method} \alias{residuals,unmarkedFitOccuFP-method} \alias{residuals,unmarkedFitOccuRN-method} \alias{residuals,unmarkedFitOccuMulti-method} \alias{residuals,unmarkedFitOccuTTD-method} \alias{residuals,unmarkedFitGDR-method} \alias{residuals,unmarkedFitIDS-method} \alias{residuals,unmarkedFitOccuCOP-method} \alias{update,unmarkedFit-method} \alias{update,unmarkedFitColExt-method} \alias{update,unmarkedFitGMM-method} \alias{update,unmarkedFitOccuMulti-method} \alias{update,unmarkedFitOccuMS-method} \alias{update,unmarkedFitOccuTTD-method} \alias{update,unmarkedFitNmixTTD-method} \alias{update,unmarkedFitGDR-method} \alias{update,unmarkedFitIDS-method} \alias{update,unmarkedFitDailMadsen-method} \alias{update,unmarkedFitGOccu-method} \alias{sampleSize} \alias{sampleSize,unmarkedFit-method} \alias{unmarkedFitOccu-class} \alias{unmarkedFitOccuPEN-class} \alias{unmarkedFitOccuPEN_CV-class} \alias{unmarkedFitOccuFP-class} \alias{unmarkedFitPCount-class} \alias{unmarkedFitDS-class} \alias{unmarkedFitMPois-class} \alias{unmarkedFitPCO-class} \alias{unmarkedFitGMM-class} \alias{unmarkedFitOccuMulti-class} \alias{unmarkedFitOccuMS-class} \alias{unmarkedFitOccuTTD-class} \alias{unmarkedFitNmixTTD-class} \alias{unmarkedFitDSO-class} \alias{unmarkedFitMMO-class} \alias{unmarkedFitIDS-class} \alias{plot,profile,missing-method} \alias{show,unmarkedFit-method} \alias{summary,unmarkedFit-method} \alias{summary,unmarkedFitDS-method} \alias{summary,unmarkedFitIDS-method} \alias{smoothed} \alias{smoothed,unmarkedFitColExt-method} \alias{projected} \alias{projected,unmarkedFitColExt-method} \alias{logLik} \alias{logLik,unmarkedFit-method} \alias{LRT} \alias{LRT,unmarkedFit,unmarkedFit-method} \alias{getY,unmarkedFit-method} \alias{getY,unmarkedFitOccu-method} \alias{getY,unmarkedFitColExt-method} \alias{getY,unmarkedFitOccuRN-method} \alias{getY,unmarkedFitOccuMulti-method} \title{Class "unmarkedFit" } \description{Contains fitted model information which can be manipulated or extracted using the methods described below. } \section{Slots}{ \describe{ \item{\code{fitType}:}{Object of class \code{"character"} } \item{\code{call}:}{Object of class \code{"call"} } \item{\code{formula}:}{Object of class \code{"formula"} } \item{\code{data}:}{Object of class \code{"unmarkedFrame"} } \item{\code{sitesRemoved}:}{Object of class \code{"numeric"} } \item{\code{estimates}:}{Object of class \code{"unmarkedEstimateList"} } \item{\code{AIC}:}{Object of class \code{"numeric"} } \item{\code{opt}:}{Object of class \code{"list"} containing results from \code{\link{optim}} } \item{\code{negLogLike}:}{Object of class \code{"numeric"} } \item{\code{nllFun}:}{Object of class \code{"function"} } \item{\code{knownOcc}:}{unmarkedFitOccu only: sites known to be occupied} \item{\code{K}:}{unmarkedFitPCount only: upper bound used in integration} \item{\code{mixture}:}{unmarkedFitPCount only: Mixing distribution} \item{\code{keyfun}:}{unmarkedFitDS only: detection function used by \link{distsamp}} \item{\code{unitsOut}:}{unmarkedFitDS only: density units} } } \section{Methods}{ \describe{ \item{[}{\code{signature(x = "unmarkedFit", i = "ANY", j = "ANY", drop = "ANY")}: extract one of names(obj), eg 'state' or 'det' } \item{backTransform}{\code{signature(obj = "unmarkedFit")}: back-transform parameters to original scale when no covariate effects are modeled } \item{coef}{\code{signature(object = "unmarkedFit")}: returns parameter estimates. type can be one of names(obj), eg 'state' or 'det'. If altNames=TRUE estimate names are more specific. } \item{confint}{\code{signature(object = "unmarkedFit")}: Returns confidence intervals. Must specify type and method (either "normal" or "profile") } \item{fitted}{\code{signature(object = "unmarkedFit")}: returns expected values of Y } \item{getData}{\code{signature(object = "unmarkedFit")}: extracts data } \item{getP}{\code{signature(object = "unmarkedFit")}: calculates and extracts expected detection probabilities } \item{getFP}{\code{signature(object = "unmarkedFit")}: calculates and extracts expected false positive detection probabilities } \item{getB}{\code{signature(object = "unmarkedFit")}: calculates and extracts expected probabilities a true positive detection was classified as certain } \item{hessian}{\code{signature(object = "unmarkedFit")}: Returns hessian matrix } \item{linearComb}{\code{signature(obj = "unmarkedFit", coefficients = "matrixOrVector")}: Returns estimate and SE on original scale when covariates are present } \item{mle}{\code{signature(object = "unmarkedFit")}: Same as coef(fit)? } \item{names}{\code{signature(x = "unmarkedFit")}: Names of parameter levels } \item{nllFun}{\code{signature(object = "unmarkedFit")}: returns negative log-likelihood used to estimate parameters } \item{parboot}{\code{signature(object = "unmarkedFit")}: Parametric bootstrapping method to assess goodness-of-fit } \item{plot}{\code{signature(x = "unmarkedFit", y = "missing")}: Plots expected vs. observed values } \item{predict}{\code{signature(object = "unmarkedFit")}: Returns predictions and standard errors for original data or for covariates in a new data.frame } \item{profile}{\code{signature(fitted = "unmarkedFit")}: used by confint method='profile' } \item{residuals}{\code{signature(object = "unmarkedFit")}: returns residuals } \item{sampleSize}{\code{signature(object = "unmarkedFit")}: returns number of sites in sample } \item{SE}{\code{signature(obj = "unmarkedFit")}: returns standard errors } \item{show}{\code{signature(object = "unmarkedFit")}: concise results } \item{summary}{\code{signature(object = "unmarkedFit")}: results with more details } \item{update}{\code{signature(object = "unmarkedFit")}: refit model with changes to one or more arguments } \item{vcov}{\code{signature(object = "unmarkedFit")}: returns variance-covariance matrix } \item{smoothed}{\code{signature(object="unmarkedFitColExt")}: Returns the smoothed trajectory from a colonization-extinction model fit. Takes additional logical argument mean which specifies whether or not to return the average over sites.} \item{projected}{\code{signature(object="unmarkedFitColExt")}: Returns the projected trajectory from a colonization-extinction model fit. Takes additional logical argument mean which specifies whether or not to return the average over sites.} \item{logLik}{\code{signature(object="unmarkedFit")}: Returns the log-likelihood.} \item{LRT}{\code{signature(m1="unmarkedFit", m2="unmarkedFit")}: Returns the chi-squared statistic, degrees-of-freedom, and p-value from a Likelihood Ratio Test.} } } \note{ This is a superclass with child classes for each fit type } \examples{ showClass("unmarkedFit") # Format removal data for multinomPois data(ovendata) ovenFrame <- unmarkedFrameMPois(y = ovendata.list$data, siteCovs = as.data.frame(scale(ovendata.list$covariates[,-1])), type = "removal") # Fit a couple of models (fm1 <- multinomPois(~ 1 ~ ufc + trba, ovenFrame)) summary(fm1) # Apply a bunch of methods to the fitted model # Look at the different parameter types names(fm1) fm1['state'] fm1['det'] # Coefficients from abundance part of the model coef(fm1, type='state') # Variance-covariance matrix vcov(fm1, type='state') # Confidence intervals using profiled likelihood confint(fm1, type='state', method='profile') # Expected values fitted(fm1) # Original data getData(fm1) # Detection probabilities getP(fm1) # log-likelihood logLik(fm1) # Back-transform detection probability to original scale # backTransform only works on models with no covariates or # in conjunction with linearComb (next example) backTransform(fm1, type ='det') # Predicted abundance at specified covariate values (lc <- linearComb(fm1, c(Int = 1, ufc = 0, trba = 0), type='state')) backTransform(lc) # Assess goodness-of-fit parboot(fm1) plot(fm1) # Predict abundance at specified covariate values. newdat <- data.frame(ufc = 0, trba = seq(-1, 1, length=10)) predict(fm1, type='state', newdata=newdat) # Number of sites in the sample sampleSize(fm1) # Fit a new model without covariates (fmNull <- update(fm1, formula = ~1 ~1)) # Likelihood ratio test LRT(fm1, fmNull) } \keyword{classes}