douconca (version 1.2.3)

fitted.dcca: Fitted values of double-constrained correspondence analysis (dc-CA)

Description

Community weighted means (CWM) and species-niche centroids (SNC), as fitted (in full or reduced rank) from the environmental data and trait data, respectively, and the fitted response from trait and environment data.

Usage

# S3 method for dcca
fitted(object, ..., type = c("CWM", "SNC", "response"), rank = "full")

Value

a matrix with fitted value. The exact content of the matrix depends on the type of fits that are asked for.

Arguments

object

return value of dc_CA.

...

Other arguments passed to the function (currently ignored).

type

type of prediction, c( "CWM","SNC", "response") for environmental values, values of traits, response (expected abundance).

rank

rank (number of axes to use). Default "full" for all axes (no rank-reduction).

Details

If type="response" the rowsums of object$data$Y are used to scale the fit to these sums. Many of the predicted response values may be negative, indicating expected absences (0) or small expected response values.

Examples

data("dune_trait_env")

# rownames are carried forward in results
rownames(dune_trait_env$comm) <- dune_trait_env$comm$Sites

mod <- dc_CA(formulaEnv = ~ A1 + Moist + Mag + Use + Condition(Manure),
             formulaTraits = ~ SLA + Height + LDMC + Condition(Seedmass) + Lifespan,
             response = dune_trait_env$comm[, -1],  # must delete "Sites"
             dataEnv = dune_trait_env$envir,
             dataTraits = dune_trait_env$traits, 
             verbose = FALSE)
# fit the mean traits at each site (20x6),
# that is CWM at each site
CWM.traits <- fitted(mod, type = "CWM")
head(CWM.traits)

# fit the mean environment for each species (28x8)
# that is SNC of each species
SNC.env <- fitted(mod, type = "SNC")
head(SNC.env)

fit.resp <- fitted(mod, type = "response")
# fitted often gives negative values and dc_CA cannot have negatives in the 
# response so, modify fit.resp,
# which gives about similar eigenvalues as the original data
fit.resp[fit.resp < 0] <- 0
mod3 <- dc_CA(formulaEnv = mod$formulaEnv,
              formulaTraits = mod$formulaTraits,
              response = fit.resp, 
              dataEnv = dune_trait_env$envir,
              dataTraits = dune_trait_env$traits, 
              verbose = FALSE)
mod3$eigenvalues / mod$eigenvalues