Merge pull request #2435 from stan-dev/feature/2336-manual-2.17

updated manual for 2.17, fixes #2336

src/docs/bibtex/all.bib

@@ -1,6 +1,16 @@
 % a do-nothing command that serves a purpose
 @preamble{ " \newcommand{\noop}[1]{} " }
 
+@article{lancaster:2000,
+  title={The incidental parameter problem since 1948},
+  author={Lancaster, Tony},
+  journal={Journal of Econometrics},
+  volume={95},
+  number={2},
+  pages={391--413},
+  year={2000}
+}
+
 @manual{minpack:1980,
 	Address = {9700 South Cass Avenue, Argonne, Illinois 60439},
 	Author = {Jorge J. More, Burton S. Garbow, Kenneth E. Hillstrom},
@@ -36,7 +46,7 @@ @article{AhnertMulansky:2011
   volume={1110.3397}
 }
 
-@article{DormandPrince:1980, 
+@article{DormandPrince:1980,
   author={Dormand, John R and Prince, Peter J},
   year={1980},
   title={A family of embedded {R}unge-{K}utta formulae},
@@ -67,11 +77,11 @@ @inproceedings{SerbanHindmarsh:2005
 
 @article{HoetingEtAl:1999,
   title={Bayesian model averaging: a tutorial},
-  author={Hoeting, Jennifer A. and Madigan, David 
+  author={Hoeting, Jennifer A. and Madigan, David
           and Raftery, Adrian E and Volinsky, Chris T.},
   journal={Statistical Science},
   volume = {14},
-  number = {4}, 
+  number = {4},
   pages={382--417},
   year={1999}
 }
@@ -1251,4 +1261,3 @@ @article{ZouHastie:2005
   number = {2},
   pages = {301--320}
 }
-

src/docs/stan-reference/acknowledgements.tex

@@ -103,6 +103,7 @@ \subsection*{Code  and Doc Patches}
 Jacob Egner,
 Ashley Ford,
 Jan Gl\"ascher,
+Jan Gleixner,
 Robert J.\ Goedman,
 Danny Goldstein,
 Tom Haber,
@@ -139,9 +140,11 @@ \subsection*{Code  and Doc Patches}
 
 Thanks for documentation bug reports and patches to:
 alvaro1101 (GitHub handle),
+andrasm (GitHub handle),
 Avraham Adler,
 Chris Anderson,
 Asim,
+Jens Astrom,
 Jarret Barber,
 Ryan Batt,
 Frederik Beaujean,
@@ -152,13 +155,16 @@ \subsection*{Code  and Doc Patches}
 Portia Brat,
 Arthur Breitman,
 Eric C.~Brown,
+Eric N.~Brown,
 Juan Sebasti\'an Casallas,
 Alex Chase,
 Daniel Chen,
 Roman Cheplyaka,
 Andy Choi,
 David Chudzicki,
 Michael Clerx,
+Marco Colombo,
+Luis Damiano,
 Andria Dawson,
 daydreamt (GitHub handle),
 Conner DiPaolo,
@@ -175,6 +181,7 @@ \subsection*{Code  and Doc Patches}
 Mauricio Garnier-Villarreal,
 Christopher Gandrud,
 Jonathan Gilligan,
+Aaron Goodman,
 John Hall,
 David Hallvig,
 David Harris,
@@ -205,15 +212,18 @@ \subsection*{Code  and Doc Patches}
 Tamas Papp,
 Anders Gorm Pedersen,
 Tomi Peltola,
+Alex Perrone,
 Andre Pfeuffer,
 Sergio Polini,
 Joerg Rings,
 Sean O'Riordain,
 Brendan Rocks,
+David M.\ Rogers,
 Cody Ross,
 Mike Ross,
 Tony Rossini,
 Nathan Sanders,
+Massimo Santini,
 James Savage,
 Terrance Savitsky,
 Dan Schrage,
@@ -222,23 +232,27 @@ \subsection*{Code  and Doc Patches}
 Janne Sinkkonen,
 skanskan (GitHub handle),
 Yannick Spill,
+Trey Spiller,
 sskates (GitHub handle),
+ssp3nc3r (GitHub handle),
 Martin Stjernman,
 Dan Stowell,
 Alexey Stukalov,
 Dougal Sutherland,
 John Sutton,
 Maciej Swat,
+Jonathan Sweeney,
 J.~Takoua,
 Andrew J.~Tanentzap,
 Shravan Vashisth,
 Aki Vehtari,
+Lukas Vermeer,
 Damjan Vukcevic,
 Matt Wand,
 Amos Waterland,
 Sebastian Weber,
 Sam Weiss,
-Luke Wiklendt,
+Lukasz Wiklendt,
 wrobell (GitHub handle),
 Howard Zail,
 Jon Zelner, and

src/docs/stan-reference/algorithms.tex

@@ -53,7 +53,7 @@ \subsection{Auxiliary Momentum Variable}
 for details of the geometry.
 
 In Stan, this matrix may be set to the identity matrix (i.e., unit
-diagonal) or estimated from warmup samples and optionally restricted
+diagonal) or estimated from warmup draws and optionally restricted
 to a diagonal matrix. The inverse $\Sigma^{-1}$ is known as the mass
 matrix, and will be a unit, diagonal, or dense if $\Sigma$ is.
 
@@ -384,7 +384,7 @@ \subsection{Discretization-Interval Adaptation Parameters}
 By setting the target acceptance parameter $\delta$ to a value closer
 to 1 (its value must be strictly less than 1 and its default value is
 0.8), adaptation will be forced to use smaller step sizes. This can
-improve sampling efficiency (effective samples per iteration) at the
+improve sampling efficiency (effective sample size per iteration) at the
 cost of increased iteration times. Raising the value of $\delta$ will
 also allow some models that would otherwise get stuck to overcome
 their blockages.
@@ -729,15 +729,15 @@ \section{Divergent Transitions}
 random walk and biasing estimates by not being able to thoroughly
 explore the posterior distribution.  \cite{Betancourt:2016b} provides
 details of the theory, computation, and practical implications of
-divergent transitions in Hamiltonian Monte Carlo.  
+divergent transitions in Hamiltonian Monte Carlo.
 
 The Stan interfaces report divergences as warnings and provide ways to
 access which iterations encountered divergences.  ShinyStan provides
 visualizations that highlight the starting point of divergent
 transitions to diagnose where the divergences arise in parameter
 space.  A common location is in the neck of the funnel in a centered
 parameterization (as in the funnel example discussed in
-\refsection{reparameterization}). 
+\refsection{reparameterization}).
 
 If the posterior is highly curved, very small step sizes are required
 for this gradient-based simulation of the Hamiltonian to be accurate.
@@ -1341,7 +1341,7 @@ \section{Unit Vector}\label{unit-vector.section}
 if it has unit Euclidean length, so that
 %
 \[
-\Vert x \Vert 
+\Vert x \Vert
 \ = \ \sqrt{x^{\top}\,x}
 \ = \ \sqrt{x_1^2 + x_2^2 + \cdots + x_n^2}
 \ = \ 1\ .
@@ -2155,15 +2155,15 @@ \section{Stochastic Gradient Ascent}
 \subsection{Monte Carlo Approximation of the ELBO}
 
 ADVI uses Monte Carlo integration to approximate the variational
-objective function, the ELBO. The number of samples used to
+objective function, the ELBO. The number of draws used to
 approximate the ELBO is denoted by \texttt{elbo\_samples}. We
 recommend a default value of $100$, as we only evaluate the ELBO every
 \texttt{eval\_elbo} iterations, which also defaults to $100$.
 
 \subsection{Monte Carlo Approximation of the Gradients}
 
 ADVI uses Monte Carlo integration to approximate the gradients of the
-ELBO. The number of samples used to approximate the gradients is
+ELBO. The number of draws used to approximate the gradients is
 denoted by \texttt{grad\_samples}. We recommend a default value of
 $1$, as this is the most efficient. It also a very noisy estimate of
 the gradient, but stochastic gradient ascent is capable of following
@@ -2256,5 +2256,3 @@ \section{Speed Warning and Data Trimming}
 very long time, especially in models with latent parameters that grow
 with the data size. It can be helpful to diagnose a model with smaller
 data sizes in such cases.
-
-

src/docs/stan-reference/appendices.tex

@@ -108,7 +108,7 @@ \chapter{Stan for Users of BUGS}\label{stan-for-bugs.appendix}
 Bugs.
 
 To start, take a look at the files of translated \BUGS models at
-\url{http://mc-stan.org/}.  These are 40 or so models from the \BUGS
+\url{http://mc-stan.org}.  These are 40 or so models from the \BUGS
 example volumes, all translated and tested (to provide the same
 answers as \BUGS) in Stan.  For any particular model you want to fit,
 you can look for similar structures in these examples.
@@ -526,13 +526,13 @@ \section{Some Differences when Running from R}
 
 \item Stan can be set up from within R using two lines of code.
   Follow the instructions for running Stan from R on
-  \url{http://mc-stan.org/}.  You don't need to separately download
+  \url{http://mc-stan.org}.  You don't need to separately download
   Stan and RStan.  Installing RStan will automatically set up Stan.
   When RStan moves to CRAN, it will get even easier.
 \item In practice we typically run the same Stan model repeatedly.  If
   you pass RStan the result of a previously fitted model the model will
   not need be recompiled. An example is given on the running
-  Stan from R pages available from \code{http://mc-stan.org/}.
+  Stan from R pages available from \code{http://mc-stan.org}.
 \item When you run Stan, it saves various conditions including
   starting values, some control variables for the tuning and running
   of the no-U-turn sampler, and the initial random seed. You can
@@ -560,7 +560,7 @@ \section{The Stan Community}
 \begin{itemize}
 \item Stan, like WinBUGS, OpenBUGS, and JAGS, has an active community,
   which you can access via the user's mailing list and the developer's
-  mailing list; see \code{http://mc-stan.org/} for information on
+  mailing list; see \code{http://mc-stan.org} for information on
   subscribing and posting and to look at archives.
 \end{itemize}
 
@@ -663,7 +663,7 @@ \subsection{Expressions}
 \small
 \begin{Verbatim}[fontsize=\small]
 expressions ::= expression % ','
-expression ::= numeric_literal
+expression ::= real_literal
              | variable
              | '{' expressions '}'
              | expression `?` expression `:` expression
@@ -687,8 +687,6 @@ \subsection{Expressions}
 
 indexes ::= index % ','
 
-numeric_literal ::= integer_literal | real_literal
-
 integer_literal ::= 0 | [1-9] [0-9]*
 
 real_literal ::= integer_literal ?('.' [0-9]*) ?exp_literal
@@ -715,8 +713,8 @@ \subsection{Statements}
    | 'target' '+=' expression
    | 'break'
    | 'continue'
-   | 'print' '(' (expression | string_literal)% ',' ')'
-   | 'reject' '(' (expression | string_literal)% ',' ')'
+   | 'print' '(' (expression | string_literal) % ',' ')'
+   | 'reject' '(' (expression | string_literal) % ',' ')'
    | 'return' expression
    | ''
 

src/docs/stan-reference/contributed.tex

@@ -3,7 +3,7 @@ \part{Contributed Modules}
 \chapter{Contributed Modules}
 
 \noindent
-Stan is an open-source project and welcomes user contributions.  
+Stan is an open-source project and welcomes user contributions.
 
 In order to reduce maintenance on the main trunk of Stan development
 and to allow developer-specified licenses, contributed Stan modules
@@ -16,10 +16,10 @@ \section{Contributing a Stan Module}
 Stan developers either through one of the following.
 %
 \begin{itemize}
-\item \code{stan-users} mailing list: 
+\item Stan forums:
 \\
-\url{https://groups.google.com/forum/?fromgroups#!forum/stan-users}
-\item Stan e-mail: 
+\url{http://discourse.mc-stan.org}
+\item Stan e-mail:
 \\
 \href{mailto:[email protected]}{\tt [email protected]}
 \end{itemize}
@@ -58,5 +58,3 @@ \subsection{Emacs Stan Mode}
 {\it Authors:} & Jeffrey Arnold, Daniel Lee
 \end{tabular}
 \end{quote}
-
-

src/docs/stan-reference/distributions.tex

@@ -118,7 +118,7 @@ \section{Cumulative Distribution Functions}
 \ = \
 \mbox{Pr}[Y < y]
 \ = \
-\int_{-\infty}^y p(y \, | \, \theta) \ \mathrm{d}\theta.
+\int_{-\infty}^y p(y \, | \, \theta) \ \mathrm{d}y.
 \]
 The complementary cumulative distribution function (CCDF) is defined
 as
@@ -2931,8 +2931,3 @@ \subsubsection{Stan Functions}
 freedom \farg{nu} and symmetric and positive-definite scale matrix
  \farg{Sigma}; may only be used in generated quantities block}
 \end{description}
-
-
-
-
-