updated docs

DESCRIPTION

@@ -1,11 +1,11 @@
 Package: gamlr  
 Title:  Gamma Lasso Regression  
 Version: 1.13-4  
-Author: Matt Taddy <taddy@chicagobooth.edu>  
-Maintainer: Matt Taddy <taddy@chicagobooth.edu>  
+Author: Matt Taddy <mataddy@gmail.com>  
+Maintainer: Matt Taddy <mataddy@gmail.com>  
 Depends: R (>= 2.15), Matrix, methods, graphics, stats
 Suggests: parallel    
-Description: The gamma lasso algorithm provides regularization paths corresponding to a range of non-convex cost functions between L0 and L1 norms.  As much as possible, usage for this package is analogous to that for the glmnet package (which does the same thing for penalization between L1 and L2 norms).  For details see: Taddy (2015), One-Step Estimator Paths for Concave Regularization, http://arxiv.org/abs/1308.5623.
+Description: The gamma lasso algorithm provides regularization paths corresponding to a range of non-convex cost functions between L0 and L1 norms.  As much as possible, usage for this package is analogous to that for the glmnet package (which does the same thing for penalization between L1 and L2 norms).  For details see: Taddy (2016), One-Step Estimator Paths for Concave Regularization, http://arxiv.org/abs/1308.5623.
 License: GPL-3  
-URL: http://github.com/TaddyLab/gamlr, http://faculty.chicagobooth.edu/matt.taddy/index.html   
+URL: http://github.com/TaddyLab/gamlr
 

man/AICc.Rd

@@ -16,7 +16,7 @@ AICc(object, k=2)
   A numeric value for every model evaluated.
 }
 \author{
-  Matt Taddy \email{taddy@chicagobooth.edu}
+  Matt Taddy \email{mataddy@gmail.com}
 }
 \references{Hurvich, C. M. and C-L Tsai, 1989.  "Regression and Time Series Model Selection in Small Samples", Biometrika 76.}
 

man/gamlr.Rd

@@ -112,11 +112,11 @@ to the same domain as \code{y}.}
   \item{family}{The exponential family model.}
 }
 \author{
-  Matt Taddy \email{taddy@chicagobooth.edu}
+  Matt Taddy \email{mataddy@gmail.com}
 }
 \note{Under \code{prexx=TRUE} (requires \code{family="gaussian"}), weighted covariances \eqn{(VX)'X} and \eqn{(VX)'y}, weighted column sums of \eqn{VX}, and column means \eqn{\bar{x}} will be pre-calculated. Here \eqn{V} is the diagonal matrix of least squares weights (\code{obsweights}, so \eqn{V} defaults to \eqn{I}).  It is not necessary (they will be built by \code{gamlr} otherwise), but you have the option to pre-calculate these sufficient statistics yourself as arguments \code{vxx} (\code{matrix} or \code{dspMatrix}), \code{vxy}, \code{vxsum}, and \code{xbar} (all \code{vectors}) respectively.  Search \code{PREXX} in \code{gamlr.R} to see the steps involved, and notice that there is very little argument checking -- do at your own risk.  Note that \code{xbar} is an \emph{unweighted} calculation, even if \eqn{V \neq I}.   For really Big Data you can then run with \code{x=NULL} (e.g., if these statistics were calculated on distributed machines and full design is unavailable). \emph{Beware:} in this \code{x=NULL} case our deviance (and df, if \code{gamma>0}) calculations are incorrect and selection rules will always return the smallest-lambda model.
 }
-\references{Taddy (2013), The Gamma Lasso, http://arxiv.org/abs/1308.5623}
+\references{Taddy (2016), One-Step Estimator Paths for Concave Regularization, Journal of Computational and Graphical Statistics, http://arxiv.org/abs/1308.5623}
 
 \examples{
 

man/hockey.Rd

@@ -13,8 +13,7 @@
   See the Chicago hockey analytics project at \code{github.com/mataddy/hockey}.}
 \value{
    \item{goal}{ Info about each goal scored, including \code{homegoal} -- an indicator for the home team scoring.}
-   \item{player}{ Sparse Matrix with entries for who was on the ice for each goal: +1 for a home team player, -1 for an away
-	 team player, zero otherwise. }
+   \item{player}{ Sparse Matrix with entries for who was on the ice for each goal: +1 for a home team player, -1 for an away team player, zero otherwise. }
    \item{team}{ Sparse Matrix with indicators for each team*season interaction: +1 for home team, -1 for away team. }
    \item{config}{ Special teams info. For example,
    \code{S5v4} is a 5 on 4 powerplay,
@@ -23,12 +22,13 @@
  }
 
 \references{Gramacy, Jensen, and Taddy (2013): "Estimating Player
-  Contribution in Hockey with Regularized Logistic Regression."
-  http://arxiv.org/abs/1209.5026.
+  Contribution in Hockey with Regularized Logistic Regression", the Journal of Quantitative Analysis in Sport.
+
+  Gramacy, Taddy, and Tian (2015): "Hockey Player Performance via Regularized Logistic Regression", the Handbook of statistical methods for design and analysis in sports.
 }
 
 \author{
-Matt Taddy, \email{taddy@chicagobooth.edu}
+Matt Taddy, \email{mataddy@gmail.com}
 }
 
 \examples{