K.mult: Test phylogenetic signal (Kmult) using phylogenetic... In phylocurve: Phylogenetic Comparative Methods for High-Dimensional Traits

Description

An implementation of Kmult (Blomberg et al. 2003; Adams 2014) in which Kmult is compared to the null distribution generated by simulation on a star phylogeny (rather than phylogenetic permutation). Additionally, the expectation of the ratio of raw to phylogenetic mean squared error (the denominator of the formula for K) is estimated via phylogenetic simulation under Brownian motion.

Usage

 `1` ```K.mult(model, nsim = 1000, plot = TRUE) ```

Arguments

 `model` An object of class `evo.model`. `nsim` The number of iterations for phylogenetic simulation. `plot` Whether or not to plot the null and alternative distribution of Kmult.

Value

An object of class `compare.model`.

Eric W. Goolsby

References

Golsby E.W. 2015. Likelihood-based parameter estimation for high-dimensional phylogenetic comparative models: overcoming the limitations of 'distance-based' methods. In review.

Adams D.C. 2014. A generalized K statistic for estimating phylogenetic signal from shape and other high-dimensional multivariate data. Syst. Biol. 63:685-697.

`evo.model` `compare.models` `physignal`

Examples

 ```1 2 3 4 5``` ```rand.data <- sim.traits() null.model <- evo.model(tree = rand.data\$tree, Y = rand.data\$trait_data,method = "Pairwise ML") K.mult(model = null.model,nsim = 100) ```

Example output

```sh: 1: cannot create /dev/null: Permission denied
sh: 1: cannot create /dev/null: Permission denied
Warning messages:
1: In rgl.init(initValue, onlyNULL) : RGL: unable to open X11 display
2: 'rgl_init' failed, running with rgl.useNULL = TRUE
call: fun(...)

Bootstrapping under null model.

**********Simulation results**********
Test statistic (K)      1.7805348
Critical test statistic 0.4943335
Estimated Power         1.0000000

P-value: 0
```

phylocurve documentation built on Aug. 9, 2017, 9:03 a.m.