[Rsiena-commits] r90 - pkg/RSienaTest/doc
noreply at r-forge.r-project.org
noreply at r-forge.r-project.org
Sun May 30 00:23:56 CEST 2010
Author: ripleyrm
Date: 2010-05-30 00:23:55 +0200 (Sun, 30 May 2010)
New Revision: 90
Modified:
pkg/RSienaTest/doc/RSiena.bib
pkg/RSienaTest/doc/s_man400.tex
Log:
Major changes to manual and minor to bib file.
Modified: pkg/RSienaTest/doc/RSiena.bib
===================================================================
--- pkg/RSienaTest/doc/RSiena.bib 2010-05-29 21:10:40 UTC (rev 89)
+++ pkg/RSienaTest/doc/RSiena.bib 2010-05-29 22:23:55 UTC (rev 90)
@@ -704,7 +704,7 @@
@ARTICLE{KiesnerEA04,
AUTHOR="Jeff {Kiesner}, Margaret {Kerr}, H{\aa}kan {Stattin} ",
- TITLE = "��Very Important Persons'' in adolescence: going beyond
+ TITLE = "‘‘Very Important Persons'' in adolescence: going beyond
in-school, single friendships in the study of peer homophily",
YEAR = "2004",
JOURNAL = "Journal of Adolescence",
@@ -1603,12 +1603,12 @@
AUTHOR="Frans N. {Stokman}",
TITLE = "Was verbindet uns wann mit wem? Inhalt und Struktur in der Analyse sozialer Netzwerke",
YEAR = "2005",
- JOURNAL = "K�lner Zeitschrift f�r Soziologie"
+ JOURNAL = "Kölner Zeitschrift für Soziologie"
}
@INCOLLECTION{StokmanZeggelink96,
AUTHOR = "Frans N. {Stokman}, Evelien P.H. {Zeggelink}",
- TITLE = "�Self-organizing' friendship networks",
+ TITLE = "‘Self-organizing' friendship networks",
BOOKTITLE = "Frontiers in social dilemmas research",
PUBLISHER = "Springer",
YEAR = "1996",
@@ -1641,7 +1641,7 @@
Author = {ter Braak, C.J.F.},
Title = {Permutation versus Bootstrap Significance Tests in Mulitple Regression and ANOVA.},
BookTitle = {Bootstrapping and Related Techniques},
- Editor = {J�ckel, K-H. and Rothe, G. and Sendler, W.},
+ Editor = {Jöckel, K-H. and Rothe, G. and Sendler, W.},
Publisher = {Springer-Verlag},
Address = {Berlin},
Year = {1992} }
Modified: pkg/RSienaTest/doc/s_man400.tex
===================================================================
--- pkg/RSienaTest/doc/s_man400.tex 2010-05-29 21:10:40 UTC (rev 89)
+++ pkg/RSienaTest/doc/s_man400.tex 2010-05-29 22:23:55 UTC (rev 90)
@@ -241,32 +241,32 @@
and \citet*{SteglichEA06}.
A website for \SI is maintained at \url{http://www.stats.ox.ac.uk/~snijders/siena/}~.
-At this website
-(`publications' tab) you shall also find references to introductions in various other languages.
+At this website (`publications' tab)
+you shall also find references to introductions in various other languages.
-
-This is a provisional manual for \SI version 4.0,
-which is also called \rs.
-This is a contributed package for the \R statistical system
+This is a manual for \SI version 4.0,
+which is also called \rs; the manual is provisional in the sense
+that it still is continually being updated.
+\RS is a contributed package for the \R statistical system
which can be downloaded from\\
\url{http://cran.r-project.org}. For the operation of \Rn,
the reader is referred
to the corresponding manual. If desired, \SI can be operated \emph{apparently}
independently of \Rn, as is explained in Section~\ref{Gui}.
-Sometimes latest versions are available at
+\RS was programmed by Ruth Ripley and Krists Boitmanis, in collaboration with Tom Snijders.
+
+In addition to the `official' \R distribution of \rs, there is
+an additional distribution at R-Forge, which is
+a central platform for the development of \R packages
+offering facilities for source code management.
+Sometimes latest versions of \RS are available at
\url{http://r-forge.r-project.org/R/?group_id=461}
before being incorporated into the R package that can be downloaded from CRAN.
+In addition, at R-Forge there is a package RSienaTest which may include
+additional options that are still in the testing stage.
-\RS was programmed by
-Ruth Ripley and Krists Boitmanis, in collaboration with Tom Snijders.
-This manual is updated rather frequently, and it may be worthwhile
-to check now and then for updates.
-It is possible that the current version still bears some traces
-from the conversion of \SI version 3 to 4, and has (mistakenly)
-some remarks that apply to version 3 and not to 4.
-
\iffalse
The manual focuses on the use of \SI for analysing the dynamics
of directed networks. The case of non-directed networks is very similar,
@@ -762,6 +762,8 @@
The following is an example \R script,
which one may use to get started with \rs.
+The appendix contains a list of \RS functions which may be consulted
+in addition to this script.
\begin{verbatim}
#####################################GENERAL###################################
@@ -869,9 +871,9 @@
####################FROM VECTORS AND MATRICES TO SIENA OBJECTS##################
-# A number of objects need to be created in R, as preparations to letting Siena07
+# A number of objects need to be created in R, as preparations to letting siena07
# execute the estimation.
-# sienaModelCreate creates a control object which can be used as an argument for Siena07
+# sienaModelCreate creates a control object which can be used as an argument for siena07
# You can look in the RSiena help files, requested by typing ?RSiena,
# to find out about options that you may use here;
# for beginning users, only the two options mentioned below are relevant.
@@ -1303,8 +1305,8 @@
\item In Phases 1 and 3 the simulations are performed in parallel. In Phase 2,
multiple simulations are done with the same parameters, and the resulting
- statistics are averaged. The gain parameter is increased and the minimum numb
- er of iterations in phase 2 reduced to take advantage of
+ statistics are averaged. The gain parameter is increased and the
+ number of iterations in phase 2 reduced to take advantage of
the increased accuracy.
\item The parameters required to run all processes on one computer are fairly
@@ -1323,13 +1325,14 @@
see the documentation for the package \sfn{snow}.
\item Currently \RS uses sockets for inter-process communication.
-\item Each process needs a copy of the data in memory. If there is insuffient
+\item Each process needs a copy of the data in memory. If there is insufficient
memory available there will be no speed gain as too much time will be spent
paging.
\item In each iteration the main process waits until all the other processes
have finished. The overall speed is therefore that of the slowest process, and
there should be enough processors to allow them all to run at speed.
\end{enumerate}
+
\subsection{Steps for looking at results: Executing \si .}
\label{S_exec}
@@ -1344,13 +1347,14 @@
in absolute value, and that it has nearly converged if they are all
smaller than 0.2.\\
These bounds are indications only, and may be taken with a grain of
- salt.\\
-
- \smallskip
-
-\item The \textsf{Initial value of gain parameter} determines the
+ salt.
+\item In rare circumstances, when the data set leads to instability
+ of the algorithm, the following may be of use.\\
+ The \textsf{Initial value of the gain parameter} determines the
step sizes in the parameter updates in the iterative
algorithm.
+ This is the parameter called \textsf{firstg}
+ in function \textsf{sienaModelCreate}.
A too low value implies that it takes very long to attain a
reasonable parameter estimate when starting from an initial
parameter value that is far from the `true' parameter estimate.
@@ -1876,11 +1880,16 @@
In the current implementation of \si, missing data are treated in
a simple way, trying to minimize their influence on the estimation
results.
-This method is further explained in \citet{HuismanSteglich08},
-where comparisons are also made with other ways of dealings with the missing
-information.
+%This method is further explained in \citet{HuismanSteglich08},
+%where comparisons are also made with other ways of dealings with the missing
+%information.
The basic idea is the following.
+\medskip
+
+NOTE: This may not be a correct representation. To be modified.
+\medskip
+
A brief sketch of the procedure is that
missing values are imputed to allow meaningful simulations;
for the calculation of the target statistics in the Method of Moments,
@@ -1900,7 +1909,7 @@
if there is an earlier observed value of this variable then
the last observed value is used to impute the current
value (the `last observation carry forward' option,
-cf.\ \citet{Lepkowski89}; if there is no earlier observed
+cf.\ \citet{Lepkowski89}); if there is no earlier observed
value, the value 0 is imputed.
For the dependent behavior variables the same principle
is used: if there is a previous observation of the same variable
@@ -4339,14 +4348,14 @@
\newpage
\subsection{Testing differences between independent groups}
-Sometimes it is interesting to test differences between parameters
-estimated for independent groups. For example, for work-related support networks
-analyzed in two different firms, one might wish to test whether the
-tendency to reciprocation of work-related support, as reflected by the reciprocity parameter,
-is equally strong in both firms.
-Such a comparison is meaningful especially if the total model is the same in
-both groups, as control for different other effects would compromise
-the basis of comparison of the parameters.
+Sometimes it is interesting to test differences between parameters estimated for
+independent groups. For example, for work-related support networks analyzed in
+two different firms, one might wish to test whether the tendency to
+reciprocation of work-related support, as reflected by the reciprocity
+parameter, is equally strong in both firms. Such a comparison is meaningful
+especially if the total model is the same in both groups, as control for
+different other effects would compromise the basis of comparison of the
+parameters.
If the parameter estimates in the two networks are $\hat\beta_a$ and $\hat\beta_b$,
with standard errors \textit{s.e}$_a$ and \textit{s.e}$_b$, respectively,
@@ -4359,7 +4368,11 @@
\newpage
\subsection{Testing time heterogeneity in parameters}
-We initially assume that $\beta$ does not vary over time, yielding a \emph{restricted model}. Our data contains $|\mathcal{M}|$ observations, and we estimate the restricted model the method of moments. We wish to test whether the \emph{restricted model} is misspecified with respect to time heterogeneity. Formally, define a vector of time dummy terms $\mathbf{h}$:
+We initially assume that $\beta$ does not vary over time, yielding a
+\emph{restricted model}. Our data contains $|\mathcal{M}|$ observations, and we
+estimate the restricted model the method of moments. We wish to test whether the
+\emph{restricted model} is misspecified with respect to time
+heterogeneity. Formally, define a vector of time dummy terms $\mathbf{h}$:
\begin{align}
h_k^{(m)}=\left\{
\begin{array}{ll}
@@ -4368,22 +4381,29 @@
\end{array}
\right . ,
\end{align}
-where $k$ corresponds to an effect included in the model.\footnote{The dummy $\delta_k^{(1)}$ is always zero so that period $w_1$ is (arbitrarily) considered the reference period.} An \emph{unrestricted model} which allows for time heterogeneity in all of the effects is considered as a modification of \eqref{eq:fij}:
+where $k$ corresponds to an effect included in the model.\footnote{The dummy
+ $\delta_k^{(1)}$ is always zero so that period $w_1$ is (arbitrarily)
+ considered the reference period.} An \emph{unrestricted model} which allows
+for time heterogeneity in all of the effects is considered as a modification of
+\eqref{eq:fij}:
\begin{align}
-f^{(m)}_{ij}(\mathbf{x})= \sum_k (\beta_k + \delta_k^{(m)} h_k^{(m)}) s_{ik}(\mathbf{x}(i \leadsto j))
+f^{(m)}_{ij}(\mathbf{x})= \sum_k (\beta_k + \delta_k^{(m)} h_k^{(m)})
+s_{ik}(\mathbf{x}(i \leadsto j))
\label{eq:fmij}
\end{align}
-where $\delta_k^{(m)}$ are the time dummy interacted effect parameters. One way to formulate the testing problem of assessing time hetergeneity is the following:
+where $\delta_k^{(m)}$ are the time dummy interacted effect parameters. One way
+to formulate the testing problem of assessing time hetergeneity is the
+following:
\begin{align}
H_0:\delta_k^{(m)} = 0 & \mbox{~for all~} k,m \notag\\
H_1:\delta_k^{(m)} \neq 0 & \mbox{ for some } k,m .
\label{hyptest}
\end{align}
-An application of the score test is given for the special case of parameter heterogeneity
-by \citet{Lospinoso2010a} and implemented in RSiena.
-To apply the test to your dataset, run an estimation in the usual way, e.g.:
+An application of the score test is given for the special case of parameter
+heterogeneity by \citet{Lospinoso2010a} and implemented in RSiena. To apply the
+test to your dataset, run an estimation in the usual way, e.g.:
\begin{verbatim}
mymodel <- sienaModelCreate(fn=simstats0c, nsub=2, n3=100)
mynet1 <- sienaNet(array(c(s501, s502, s503), dim=c(50, 50, 3)))
@@ -6937,24 +6957,28 @@
% \maketitle
% \tableofcontents
% \listoftables
+
\appendix
\newpage
\section{List of Functions in Order of Execution}
This appendix, for which we are indebted to Paulina Preciado Lopez,
provides a description of the functions that constitute the
- RSiena package. This is intended as a quick reference or catalogue for the
- user to employ Stochastic Actor Oriented Models (SAOM) to analyse network
- dynamics in R.
+ \RS package. This is intended as a quick reference or catalogue for the
+ user to employ Stochastic Actor Oriented Models (SAOM) to analyze network
+ dynamics in \Rn.
- The functions are presented in execution order (in the approximate order in
- which they would be used in real model estimation) A list of useful R
+ The functions are presented in execution order (more or less as
+ they would be used in practice). A list of useful \R
functions to read and prepare the data set is also included at the
beginning. In all cases examples on how to use these functions are provided.
+ In the `syntax' column,
+ when arguments of functions are followed by = and a single option,
+ this is the default option.
- The descriptions provided are suitable for beginner and intermediate R and
+ The descriptions provided are suitable for beginner and intermediate \R and
Siena users. For the advanced specifications of the functions the user
- should refer to the help by typing ``?funName'' in the R console, where
+ should refer to the help by typing ``?funName'' in the \R console, where
``funName'' is the name of the function.
We consider that the model estimation is composed by 6 stages:
@@ -6969,101 +6993,94 @@
\end{enumerate}
Tables \ref{tab:FuncExR} and \ref{tab:ListSienaExec} present the list of useful
-R functions and the list of RSiena functions in execution order, respectively.
-\begin{footnotesize}
+\R functions and the list of \RS functions in execution order, respectively.
+%\begin{footnotesize}
\begin{sidewaystable}
\begin{threeparttable}
\centering
- \begin{tabular}{c | l | p{4.5cm} | p{5cm} | p{8cm} }
+ \begin{tabular}{c | l | p{4cm} | p{4cm} | p{12cm} }
\multicolumn{1}{c}{\textbf{Stage}} & \multicolumn{1}{c}{\textbf{Name}}
& \multicolumn{1}{c}{\textbf{Syntax}} &
\multicolumn{1}{c}{\textbf{Examples}} &
\multicolumn{1}{c}{\textbf{Description}} \\
\cline{1-5}
1 &help\tnote{*} & help(funame) & help(siena01Gui) &
- Opens the help on the function named ``funame'', this can also be done by
- typing ``?'' followed by ``funame'' in the console\\
+ Opens the help on the function named ``funame''; this can also be done by
+ typing ``?'' followed by ``funame'' in the console.
+ This is the general way to get information about further options
+ of this function. \\
\hline
1 &getwd &getwd() & &
Returns the name of the current working directory.
Does not require arguments\\
\hline
1 & list.files & list.files(dir) &
-list.files (``C:/User/ MyDocuments/ MySiena'') & Returns a character vector
-with the names of the files in the directory ``dir''.
-If no argument is provided, ``dir'' is the current working directory.
-Type ?list.files for more options\\
-\hline
-1 &setwd\tnote{*} &setwd(dir) &setwd(``C:/MyDocuments/ MySiena'') &
+list.files (``C:/User/ MyDocuments/
+\newline
+MySiena'') & Returns a character vector
+with the names of the files in the directory ``dir''. If no argument is
+provided, ``dir'' is the current working directory.\\
+\hline 1 &setwd\tnote{*} &setwd(dir) &setwd( ``C:/MyDocuments/ MySiena'') &
Sets the working directory to ``dir''. In this context the working
directory should be where the data is saved\\
-\hline
-1 &install.packages\tnote{*} & install.packages() &
-&It is used to install packages. If no arguments are provided it opens a GUI
+\hline 1 &install.packages\tnote{*} & install.packages() & &It is used to
+install packages. If no arguments are provided it opens a GUI
to select a mirror site and the packages that we want to download and
- install. This is not necessary if the package has already been installed.
- See ?install.packages for more options\\
-\hline
-1 &library\tnote{*} & library(package) & library(RSiena) &
-Loads the library named ``package''. See ?library for more options.\\
-\hline
-1 &read.table & read.table(file, header = FALSE, sep = ``'',
-quote = ``\'''',...) & net1$<$--read.table(`network1. dat' , header = F)
-& Reads a file in table format and creates a data frame from it.
-The argument ``file'' is the file containing the data. In the case of
-adjacency matrices, the file should have the same number of columns and rows.
-``header'' is a logical argument indicating whether the first row of the data
-contains the column names. ``sep'' is the field separator character
+ install. This is not necessary if the package has already been installed.\\
+\hline 1 &library\tnote{*} & library(package) & library(RSiena) &
+Loads the library named ``package''. \\
+\hline 1 &read.table & read.table(file, header=FALSE, sep=``'',
+\newline
+quote=``{\''}'',...) & net1 $<$-- read.table(
+\newline
+`network1.dat', header=F) & Reads
+a file in table format and creates a data frame from it. The argument ``file''
+is the file containing the data. In the case of adjacency matrices, the file
+should have the same number of columns and rows. ``header'' is a logical
+argument indicating whether the first row of the data contains the column
+names. ``sep'' is the field separator character
(such as space, comma, etc.). See the help on the function to specify
other arguments\\
-\hline
-2 & as.matrix &as.matrix(x,...) &net1 $<$-- as.matrix(net1) &
+\hline 2 & as.matrix &as.matrix(x,...) &net1 $<$-- \newline as.matrix(net1) &
Transforms an object ``x'' into a matrix. Siena works with matrices and
not with data frames\\
-\hline
-2& class & class(x) & class(net1) & Returns the type of object that
+\hline 2& class & class(x) & class(net1) & Returns the type of object that
``x'' is\\
\hline
2& dim & dim(x) & dim(net1) & Returns the dimension of object ``x''\\
-\hline
-4 & fix\tnote{*} & fix(x) & fix(effects) & Allows editing the object ``x''
+\hline 4 & fix\tnote{*} & fix(x) & fix(effects) & Allows editing the
+object ``x''
by opening a window and it replaces the old object by the edited ``x''\\
\hline
\end{tabular}
-\caption[Functions from R in order of execution]
-{Useful functions from R in execution order}
-\label{tab:FuncExR}
+\caption[Functions from \R in order of execution] {Useful functions from \R in
+execution order} \label{tab:FuncExR}
\begin{tablenotes}
\item [*] Also available via a menu option
\end{tablenotes}
\end{threeparttable}
\end{sidewaystable}
-\end{footnotesize}
+%\end{footnotesize}
\begin{landscape}
-\begin{footnotesize}
+%\begin{footnotesize}
\begin{longtable}{c | p{3cm} | p{5.2cm} | p{4.2cm} | p{8.5cm} }
-\caption[List of RSiena Functions: Execution]
-{List of RSiena Functions in order of Execution}
+\caption[List of \RS Functions: Execution] {List of \RS Functions in order of
+Execution}
\label{tab:ListSienaExec} \\
\hline
-\multicolumn{1}{c}{\textbf{Stage}} &
-\multicolumn{1}{c}{\textbf{Name}} &
-\multicolumn{1}{c}{\textbf{Syntax}} &
-\multicolumn{1}{c}{\textbf{Examples}} &
+\multicolumn{1}{c}{\textbf{Stage}} & \multicolumn{1}{c}{\textbf{Name}} &
+\multicolumn{1}{c}{\textbf{Syntax}} & \multicolumn{1}{c}{\textbf{Examples}} &
\multicolumn{1}{c}{\textbf{Description}} \\
\hline
\endfirsthead
\multicolumn{5}{c}%
{{\bfseries \tablename\ \thetable{} -- continued from previous page}} \\
-\hline
-\multicolumn{1}{c}{\textbf{Stage}} &
-\multicolumn{1}{c}{\textbf{Name}} &
-\multicolumn{1}{c}{\textbf{Syntax}} &
-\multicolumn{1}{c}{\textbf{Examples}} &
+\hline \multicolumn{1}{c}{\textbf{Stage}} & \multicolumn{1}{c}{\textbf{Name}} &
+\multicolumn{1}{c}{\textbf{Syntax}} & \multicolumn{1}{c}{\textbf{Examples}} &
\multicolumn{1}{c}{\textbf{Description}} \\
\hline
\endhead
@@ -7072,72 +7089,87 @@
\hline
\endfoot
-\hline
-\hline
+\hline \hline
\endlastfoot
1 & installGui & installGui() &
- &Starts the installer for the standalone version of RSiena.
+ &Starts the installer for the standalone version of \rs.
Only for Windows. Does not require arguments\\
\hline
-3 to 5& siena01Gui& siena01Gui()& & Does not require arguments.
+3 -- 5& siena01Gui& siena01Gui()& & Does not require arguments.
Opens a GUI to be used to run the model estimation or to create a session
- from which to work within R. Details on how to run the estimation under
+ from which to work within \Rn. Details on how to run the estimation under
the GUI can be found in section \ref{thegui} and \ref{estgui}.\\
\hline
-3 &sienaNodeSet &sienaNodeSet (n, nodeSetName = ``Actors'',
-names = NULL) & & Creates a Siena node set which can be used as the nodes
+3 &sienaNodeSet &sienaNodeSet (n, \newline
+nodeSetName= ``Actors'', \newline
+names=NULL) & &
+Creates a Siena node set which can be used as the nodes
in a siena network. ``n'' is the number of actors or nodes; ``nodeSetName''
is a character string to name the node set (defaults to ``Actors'') and
```names'' is a string vector with length n with the names of each node
(optional)\\
\hline
-3 & sienaNet & sienaNet (netarray, type = c(``oneMode'', ``bipartite'',
-``behavior''), nodeSet = ``Actors'', sparse = is.list (netarray)) &
-sienaNet(array(c(net1,net2, net3), dim = c(dim(net1),3))) & Creates a Siena
-network object by forming an array of network observations represented as
-matrices, arrays or sparse matrices. ``netarray'' is a matrix (type=``behavior''
-only) or array of values or list of sparse matrices of type
-``dgTMatrix'';``type'' is either ``one mode'' (default), ``bipartite'' or
-``behaviour''; ``nodeSet'' is the name of the node set. It is a vector with two
-strings for a bipartite network; ``sparse'' is logical and it is set to TRUE if
-the data is in sparse matrix
+3 & sienaNet & sienaNet (netarray, type= \newline
+c(``oneMode'', ``bipartite'',\newline
+``behavior''), \newline
+nodeSet=``Actors'', \newline
+sparse=is.list (netarray)) & sienaNet(array(
+c(net1,net2,net3), dim=c(dim(net1),3))) & Creates a Siena network object by
+forming an array of network observations represented as matrices, arrays or
+sparse matrices. ``netarray'' is a matrix (type=``behavior'' only) or array of
+values or list of sparse matrices of type ``dgTMatrix'';``type'' is either
+``one mode'' (default), ``bipartite'' or ``behaviour''; ``nodeSet'' is the name
+of the node set. It is a vector with two strings for a bipartite network;
+``sparse'' is logical and it is set to TRUE if the data is in sparse matrix
format, FALSE otherwise\\
\hline
-3 &coCovar & coCovar(val, nodeSet =`Actors') & cons $<$-- as.matrix( read.table
-('cons.DAT')) \newline cons1 $<$-- coCovar (cons[,1]) & Creates a constant
-covariate object, where val is the vector of covariate values and nodeSet is the
-name of the actors' set. The dimension of val should be (1, \#
+3 &coCovar & coCovar(val, \newline
+nodeSet =`Actors') & cons $<$-- \newline
+as.matrix( read.table \newline
+('cons.DAT')) \newline cons1 $<$--\newline
+ coCovar (cons[,1]) & Creates a constant
+covariate object, where val is the vector of covariate values and nodeSet is
+the name of the actors' set. The dimension of val should be (1, \#
Actors)\\
\hline
-3 & varCovar & varCovar(val, nodeSet =`Actors') & chan $<$-- as.matrix
-(read.table ('chan.DAT')) \newline chan $<$-- varCovar (chan[,1]) & Creates a
+3 & varCovar & varCovar(val, \newline
+nodeSet =`Actors') & chan $<$-- as.matrix \newline
+(read.table ('chan.DAT')) \newline chan $<$-- \newline
+varCovar (chan[,1]) & Creates a
changing covariate object where ``val'' is a matrix with the covariate values
with one row for each actor and one column for each period; ``nodeSet'' is the
name of the set of actors \\
\hline
-3& coDyadCovar &coDyadCovar(val, nodeSets = c(``Actors'', ``Actors'')) & &
-Creates a constant dyadic covariate object where ``val'' is a matrix of the same
-dimension as the network observations and nodeSets are the sets of actors with
+3& coDyadCovar &coDyadCovar(val, \newline
+nodeSets= \newline
+c(``Actors'', ``Actors'')) & &
+Creates a constant dyadic covariate object where ``val'' is a matrix of the
+same dimension as the network observations and nodeSets are the sets of actors
+with
which the constant covariate is associated\\
\hline
-3 &varDyadCovar & varDyadCovar(val, nodeSets = c(``Actors'', ``Actors'')) &
+3 &varDyadCovar & varDyadCovar(val, \newline
+nodeSets= \newline
+c(``Actors'', ``Actors'')) &
&Creates a changing dyadic covariate object where ``val'' is an array of
matrices. Each matrix has the same dimension of the actor set and ``val'' has
one less matrices than observations of the network; ``nodeSets'' are the sets
of actors to which the varying covariate object is associated\\
\hline
-3 & sienaComposition Change & sienaCompositionChange(changelist, nodeSet =
-"Actors", option = 1) & & Creates a list of events describing the moments in
-which each actor is present in the network: ``changelist'' is a list with an
+3 & sienaCompositionChange & \newline
+sienaCompositionChange( changelist,\newline
+nodeSet="Actors", \newline
+option=1) & & Creates a list of events describing the moments
+in which each actor is present in the network: ``changelist'' is a list with an
entry for each actor in the node set. Each entry is a vector indicating
intervals in which an actor is present in the network. ``nodeSet'' is the name
of the set of actors corresponding to these composition changes and ``option''
@@ -7146,233 +7178,289 @@
details on this\\
\hline
-3 & sienaComposition ChangeFromFile & sienaCompositionChangeFromFile (filename,
-nodeSet = "Actors", fileobj = NULL, option = 1) & & Creates a list of events
+3 & sienaCompositionChangeFromFile & \newline
+sienaCompositionChangeFromFile ( filename, \newline
+nodeSet="Actors", \newline
+fileobj=NULL, option=1) & & Creates a list of events
describing the changes over time in the actor set from a file. ``filename'' is
the name of the file containing change information (one line per actor) each
-line is a series of space delimited numbers indicating intervals. ``fileobj'' is
-the result of readLines on ``filename''. ``nodeSet'' is the name of the set of
-actors. ``option'' (defaults to 1) has the same
+line is a series of space delimited numbers indicating intervals. ``fileobj''
+is the result of readLines on ``filename''. ``nodeSet'' is the name of the set
+of actors. ``option'' (defaults to 1) has the same
description that in sienaCompositionChange\\
\hline
-3 & sienaDataCreate & sienaDataCreate(...,nodeSets=NULL, getDocumentation =
-FALSE) & MyData $<$-- sienaDataCreate (net, cons1, cons2, cons3, chan, dyad) &
-Creates a siena object from networks, covariates, composition and behaviour
-objects: .``...'' represents the objects of class ``sienaNet'', ``coCovar'',
-``varCovar'', ``coDyadCovar'', ``varDyadCovar'',
+3 & sienaDataCreate & sienaDataCreate(...,\newline nodeSets=NULL, \newline
+getDocumentation=FALSE) &
+MyData $<$-- \newline
+sienaDataCreate (net, cons1, cons2, cons3, \newline
+chan, dyad) & Creates a siena object from networks, covariates,
+composition and behaviour objects: .``...'' represents the objects of class
+``sienaNet'', ``coCovar'', ``varCovar'', ``coDyadCovar'', ``varDyadCovar'',
``compositionChange''. ``nodeSets'' is a list of Siena node sets. Default is a
-single set named ``Actors'' with length equal to the number of rows in the first
-object of class ``SienaNet'', it has to match the nodeSet supplied when the
-arguments are created; ``getDocumentation'' is a flag to allow
+single set named ``Actors'' with length equal to the number of rows in the
+first object of class ``SienaNet'', it has to match the nodeSet supplied when
+the arguments are created; ``getDocumentation'' is a flag to allow
documentation for internal functions, not for use by users\\
\hline
-3 & sienaDataCreateFrom Session & sienaDataCreateFromSession(file name = NULL,
-session = NULL, modelName = ``Siena'', ...) & myobj $<$-- sienaDataCreate
-FromSession(`Session.csv') & Reads a SIENA session from a file and creates a
+3 & sienaDataCreateFromSession & \newline
+sienaDataCreateFromSession( \newline
+filename=NULL, \newline
+session=NULL, \newline
+modelName=``Siena'', ...) & myobj $<$-- \newline
+sienaDataCreateFromSession \newline
+(`Session.csv') & Reads a SIENA session from a file and creates a
Siena Data object or group. ``file'' is the session file; ``session'' is the
input session if the function is called from siena01Gui(); ``modelName'' is the
project's name; ``...'' refers to other
arguments used by siena01Gui()\\
\hline
-3 & sienaGroupCreate & sienaGroupCreate (objlist, single OK = FALSE,
-getDocumentation=FALSE) & sienaGroupCreate (list( MyData1, MyData2)) & Creates
-an object of class ``sienaGroup'' from a list of Siena data objects: ``objlist''
-is a list of objects of class ``siena''; ``singleOK'' is a boolean variable to
-indicate if it is OK to have just one object; ``getDocumentation''
+3 & sienaGroupCreate & sienaGroupCreate (objlist, \newline
+singleOK=FALSE, \newline
+getDocumentation=FALSE) & sienaGroupCreate (list( \newline
+MyData1, MyData2)) & Creates
+an object of class ``sienaGroup'' from a list of Siena data objects:
+``objlist'' is a list of objects of class ``siena''; ``singleOK'' is a boolean
+variable to indicate if it is OK to have just one object; ``getDocumentation''
is a flag to allow documentation of internal functions, not for use by users\\
\hline
-4 & effectsDocumentation & effectsDocumentation() & & Prints a html or
+4 & effectsDocumentation & \newline
+effectsDocumentation() & & Prints a html or
\LaTeX\ table with the effects details\\
\hline
-4 & getEffects& getEffects(x, nintn = 10, behNintn = 4, getDocumentation =
-FALSE) & MyEff$<$--getEffects (MyData, nint = 2, behNint = 1) & Creates a siena
-effects objects (a data frame) that contains a list of the effects that can be
-included in the model. Type fix(MyEff) to edit the effects through a GUI
-(e.g. Including them or excluding them, changing their names, initial values,
-fixing them, etc.) The arguments are a siena or a siena group object ``x'', the
-number of lines for user defined network interactions ``nint'' and the number of
-lines for user defined behaviour interactions ``behNintn''. ``getDocumentation''
-is a flag to allow documentation for
+4 & getEffects& getEffects(x, nintn=10, \newline
+behNintn=4, \newline
+getDocumentation=FALSE) &
+MyEff $<$-- getEffects (\newline
+MyData, nint=2, \newline
+behNint=1) & Creates a siena effects
+objects (a data frame) that contains a list of the effects that can be included
+in the model. Type fix(MyEff) to edit the effects through a GUI (e.g.
+Including them or excluding them, changing their names, initial values, fixing
+them, etc.) The arguments are a siena or a siena group object ``x'', the number
+of lines for user defined network interactions ``nint'' and the number of lines
+for user defined behaviour interactions ``behNintn''. ``getDocumentation'' is a
+flag to allow documentation for
internal functions, not to be used by users\\
\hline
-4 & includeEffects & includeEffects(myeff, ..., include = TRUE, name =
-myeff\$dollar name[1], type = ``eval'', interaction1 = ``'', interaction2 =
-``'') & {MyEff$<$--includeEffects(MyEff, transTrip, balance) \flushleft
-MyEff$<$--includeEffects(MyEff, sameX, sameXRecip, interaction1="gender")}
+4 & includeEffects & includeEffects(myeff, ..., include=TRUE, \newline
+name=myeff{\,\$}name[1], type=``eval'', \newline
+interaction1=``'',\newline
+interaction2=``'') & {MyEff$<$--includeEffects(MyEff, transTrip, balance)
+\flushleft MyEff$<$--includeEffects(MyEff, sameX, \newline
+sameXRecip, \newline
+interaction1="gender")}
&The function is a
way to select the effects to be included. ``myeff'' is an effects object, as
created by getEffects. It is necessary to indicate the short names to identify
-the effects to be included (argument ...). Use myeff\$shortName to get a list of
-the short names of possible effects to include and myeff\$effectName to get the
-full name of the effects. This information can also be found in the
[TRUNCATED]
To get the complete diff run:
svnlook diff /svnroot/rsiena -r 90
More information about the Rsiena-commits
mailing list