### phylocurve features

#### Ancestral curve reconstruction

Function: ```phylocurve(formula, tree, data, ymin = 0.01, ymax = 0.99, ylength = 30, tip_coefficients, species_identifier = "species", verbose = FALSE) ```

Description: This function uses a PGLS-based method described in Goolsby (2015) to perform ancestral curve reconstruction. This function uses a fast tree transversal method via the phylolm package (Ho and Ane, 2014). Currently only logistic regression (glm with logit link) is implemented.

#### Estimation of evolutionary rates

Function: ```rate.mult(tree = tree, Y = Y, type = c("mult", "diag", "all"), method = c("REML", "ML"), error = c("none", "estimate", "supply"), error_n = 20, error_supply, model = "BM", fixed_sigma2, fixed_model_pars) ```

Description: This function estimates multivariate evolutionary rates (Adams 2014) using maximimum likelihood. The function must be run prior to running other multivariate functions, including K.mult, pgls.mult, compare.rate.mult, and compare. multivar.rate.mult.

#### Compare multivariate evolutionary rates

Function: ```compare.rate.mult(rate.mult.fitted, groups, fit_individual = FALSE) ```

Description: This function compares multivariate evolutionary rates on a tree (Adams 2014) using maximimum likelihood (O'Meara et al. 2006).

#### Compare evolutionary rates of multiple multivariate traits

Function: ```Compare multiple multivariate evolutionary rates ```

Description: Compares hypotheses about evolutionary rates for multivariate traits combining methods from O'Meara et al. (2006), Adams (2013) and Adams (2014).

#### Phylogenetic signal

Function: ```K.mult(rate.mult.fitted, iter = 1000) ```

This function estimates multivariate phylogenetic signal (Adams 2014) on a fited rate.mult object using a covariance-based approach.

#### Phylogenetic regression

Function: ```pgls.mult(rate.mult.fitted, X) ```

Description: This function performs multivariate phylogenetic regression (Adams 2014) to assess the significance of correlated evolution between a multivariate trait and a univariate trait, using a covariance-based approach.