douconca (version 1.2.1)

predict.dcca: Prediction for double-constrained correspondence analysis (dc-CA)

Description

Prediction of traits from environment, environment from traits and response from trait and environment data.

With type = "traits" and newdata = NULL, predict gives the fitted mean traits, i.e. the fitted community weighted means. With type = "env" and newdata = NULL, predict gives the fitted mean environment, i.e. the fitted species niche centroids.

Usage

# S3 method for dcca
predict(
  object,
  ...,
  type = c("env", "traits", "response", "reg_env", "reg_traits"),
  rank = "full",
  newdata = NULL
)

Value

a matrix with the predictions. The exact content of the matrix depends on the type of predictions that are being made.

Arguments

object

return value of dc_CA.

...

Other arguments passed to the function (currently ignored).

type

type of prediction, c("env", "traits", "response", "reg_env", "reg_trait") for environmental values, values of traits, response (expected abundance), regression coefficients for environmental and trait predictors.

rank

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

newdata

Data in which to look for variables with which to predict. For type = "reg_env" or "reg_trait" newdata is ignored. For type = "env" or "trait", newdata is a data frame of trait and environmental values, respectively, which are used for the prediction. If omitted, fitted values are generated. For type = "response", newdata is a list of two data frames with trait and environmental values in this order, e.g. list(traits = dataTraits, env = dataEnv).

Details

Variables that are in the model but not in newdata are set to their weighted means in the training data. Predictions are thus at the (weighted) mean of the quantitative variables not included. Predictions with not-included factors are at the weighted mean (none of the factor effects are included).

For type = "response" and non-null newdata, the species weights of the training are used; the site weights are taken equal. Many of the predicted values may be negative, indicating expected absences (0) or small expected response values.

Regression coefficients obtained with type = "reg_env" or type = "reg_traits" are for standardized traits and environmental variables.

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)

# regression coefficients
predict(mod, type = "reg_env")
predict(mod, type = "reg_traits")

# fit the mean traits at each site (20x6),
# that is CWM at each site
pred.traits <- predict(mod, type = "traits")
head(pred.traits)

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

pred.resp <- predict(mod, type = "response")
# pred has negative values and dc_CA cannot have negatives in the response
# so, modify pred.resp,
#whichgives about similar eigenvalues as the original data
pred.resp[pred.resp < 0] <- 0
mod3 <- dc_CA(formulaEnv = mod$formulaEnv,
              formulaTraits = mod$formulaTraits,
              response = pred.resp, 
              dataEnv = dune_trait_env$envir,
              dataTraits = dune_trait_env$traits)
mod3$eigenvalues / mod$eigenvalues