fbRanks (version 2.0)

coef.fbRanks: Return Coefficients from fbRanks Objects

Description

The rank.teams function outputs fbRanks objects. coef(fbRanks), where fbRanks is one's output from a rank.teams call, will return a list with vectors of coefficients for each fit and lists of coefficients for each term in a model. The fit element of a fbRanks object is a list with a fit (class "glm", "speedglm" or "glmnet") for each cluster. A cluster is a collection of teams with a path, represented by a series of games, between every team in the cluster. "attack" and "defend" are always model terms, but the user may have added on additional predictor variables (such as "surface").

Usage

"coef"(object, ...)

Arguments

object

A fbRanks object.

...

Not used.

Value

A list with elements coef.vector and coef.list. coef.vector is like the familar vector returned by a coef() call except that these vectors are held in a list where each element of the list is for the fit for a different cluster. coef.list is the vector separated into the coefficients for each model term. At the minimum, coef.list is a list with elements "attack" and "defend" because these terms always appear in the fbRanks model. Additional terms added by the user may also appear.

Examples

## Not run: 
# #This will load two dataframes: B00.scores and B00.teams
# data(B00data)
# 
# #fit a model using a particular date range for B00 teams
# #set the data range to just be the summer games of WA teams
# x=rank.teams(scores=B00.scores, teams=B00.teams, 
#              min.date="2012-5-1", max.date="2012-9-8", silent=TRUE)
# 
# #there are multiple clusters
# names(coef(x)$coef.list)
# 
# #show the coefficients for cluster 1 as a list
# #notice that fewer goals are scored on Turf
# coef(x)$coef.list[[1]]
# ## End(Not run)