Updated man page for gsl-bs to reflect predict() and its use

man/gsl-bs.Rd

@@ -55,14 +55,17 @@ basis; default is \sQuote{FALSE} }
   the examples at the end of this help file)
   \preformatted{
     B <- gsl.bs(x,degree=10)
-    }
+    B.predict <- predict(gsl.bs(x,degree=10),newx=xeval)
+  }
 
 }
 
 \value{
 
   \code{gsl.bs} returns a \code{gsl.bs} object.  A matrix of dimension
-  \sQuote{c(length(x), degree+nbreak-1)}.
+  \sQuote{c(length(x), degree+nbreak-1)}.  The generic function
+  \code{\link{predict}} extracts (or generates) predictions from the
+  returned object.
 
   A primary use is in modelling formulas to directly specify a piecewise
   polynomial term in a model. See \url{http://www.gnu.org/software/gsl/}

vignettes/crs_faq.Rnw

@@ -99,7 +99,7 @@ A BibTeX entry for LaTeX users is
   Documentation/Manuals (\url{http://cran.r-project.org/manuals.html})
   where you will discover a wealth of documentation for \texttt{R} users
   of all levels. See also the \texttt{R} task views summary page
-  (\url{http://cran.nedmirror.nl/web/views/index.html}) for
+  (\url{http://cran.r-project.org/web/views}) for
   information grouped under field of interest. A few documents that I
   mention to my students which are tailored to econometricians include
   \url{http://cran.r-project.org/doc/contrib/Verzani-SimpleR.pdf},

vignettes/spline_primer.Rnw

@@ -170,11 +170,20 @@ of $m=n+1$ terms and is given by
 where ${n\choose i}=\frac{n!}{(n-i)!i!}$, which can be expressed
 recursively as
 \begin{equation*}
-  B(x)=(1-x)\left(\sum_{i=0}^{n-1}\beta_iB_{i,n-1}(x)\right)+x\left(\sum_{i=1}^{n}\beta_iB_{i,n-1}(x)\right),
+  B(x)=(1-x)\left(\sum_{i=0}^{n-1}\beta_iB_{i,n-1}(x)\right)+x\left(\sum_{i=1}^{n}\beta_iB_{i-1,n-1}(x)\right),
 \end{equation*}
 so a degree $n$ B\'ezier curve is a linear interpolation between two
 degree $n-1$ B\'ezier curves.
 
+%% B_{i,n-1} above changed to B_{i-1,n-1} 20/1/15: "Dear Dr. Racine, I
+%% have spotted a small typo in your otherwise excellent Primer on
+%% Regression Splines
+%% (http://cran.r-project.org/web/packages/crs/vignettes/spline_primer.pdf
+%% ): on page 3, in the unnumbered equation below Eq (3) the
+%% parameters of the Bernstein polynomial in the second term should be
+%% B_{i-1,n-1} I believe (not B_{i,n-1}). Tamas
+%% (ferenci.tamas@nik.uni-obuda.hu)"
+
 \begin{example}{A quadratic B\'ezier curve as a linear interpolation
     between two linear B\'ezier curves.}
 
@@ -210,7 +219,7 @@ coefficients of the regression model.
 \begin{example}{The quadratic B\'ezier curve basis functions.}
 
   The figure below presents the bases $B_{i,n}(x)$ underlying a B\'ezier
-  curve for $i=1,\dots,2$ and $n=2$.
+  curve for $i=0,\dots,2$ and $n=2$.
 
 \begin{center}
 <<fig=TRUE,echo=FALSE>>=