envlpaster (version 0.1-2)

# simdata30nodes: A generated aster data set with 30 nodes

## Description

Simulated data for an aster analysis. Loads 7 objects.

## Usage

data(simdata30nodes)

## Format

The data frame with records for 3000 organisms over 10 years. The dataset corresponding to our aster analysis. The following four descriptions explain the elements of this dataset.
u
Indicates survival for each of the 10 years.
w
Counts offspring for each of the 10 years.
v
Indicates if w > 0 for each of the 10 years.
z
A covariate of potential interest, 10 in total.
variables
Character vector giving the names of the variables in the graph.
root
The root data. For aster.default an nind by nnode matrix, for aster.formula an nind * nnode vector.
modmat
An nind by nnode by ncoef three-dimensional array, the model matrix. aster.formula constructs such a modmat from its formula, the data frame data, and the variables in the environment of the formula.
formula
Necessary for changing to class aster.formula.
xlevels
Necessary for changing to class aster.formula.
terms
Necessary for changing to class aster.formula.
simdata30nodes.asterdata
An object of class asterdata corresponding to simdata30nodes.

## Details

This object contains an aster data set in wide form, an object of class asterdata corresponding to the original data set, and vectors specifying the graphical structure of the aster model.

There are 3000 simulated individuals in this aster analysis. Our data is generated in two parts. The first part follows Technical report 671 (TR 671) on Charlie Geyer's Aster Models for Life History Analysis webpage. For our data, nind = 3000, ntime = 10, psurv = 0.95, prepr = 0.7, mpois = 1, and the seed is set at set.seed(13) which is different from the original simulation setup.

We follow the model construction in TR 671 through out6. We then generate a new dataset from the aster model where the components of the submodel mean-value parameter vector $\tau$ corresponding to Darwinian fitness is in the space spanned by the first, second, and fourth eigenvectors of Fisher information.

## References

Geyer, C. J. and Shaw, R. G. (2009). Model Selection in Estimation of Fitness Landscapes. Technical Report No. 671. School of Statistics, University of Minnesota. http://conservancy.umn.edu/handle/11299/56219.