stepan
/
dynare
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Update of gsa documentation

time-shift
Marco Ratto 2012-06-20 15:13:29 +02:00
parent a7bddead43
commit 5539910ce4
2 changed files with 45 additions and 41 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

62
doc/gsa/gsa.tex
View File

@ -32,13 +32,13 @@
\author{Marco Ratto\\
European Commission, Joint Research Centre \\
TP361, IPSC, via E. Fermi 1\\21020 Ispra
TP361, IPSC, \\21027 Ispra
(VA) Italy\\
\texttt{marco.ratto@jrc.it}
\texttt{marco.ratto@jrc.ec.europa.eu}
\thanks{The author gratefully thanks Christophe Planas, Kenneth Judd, Michel Juillard,
Alessandro Rossi, Frank Schorfheide and the participants to the
Courses on Global Sensitivity Analysis for Macroeconomic
Models (Ispra, 2006-2007-2008) for interesting discussions and
Models (Ispra, 2006-2007-2008-2010) for interesting discussions and
helpful suggestions.}}
%%% To have the current date inserted, use \date{\today}:
@ -123,7 +123,10 @@ beforehand}.
\verb"morris_ntra"& 20& number of trajectories in Morris design\\
\verb"ppost"& 0& 0 = don't use Metropolis posterior sample\\
& & 1 = use Metropolis posterior sample: this \\
& & \hspace{0.5 cm} overrides any other sampling option! \\ \hline
& & \hspace{0.5 cm} overrides any other sampling option! \\
\verb"neighborhood_width"& []& $\delta$ (real number$>0$) uniform sample in the\\
& & neighborhood of the posterior mode $\hat{\theta}$ \\
& & interval width: $\hat{\theta}(1\pm\delta)$ \\\hline
\end{tabular}
\subsection{Stability mapping}
\begin{tabular}{r|l|l}
@ -132,11 +135,13 @@ beforehand}.
& & 0 = no stability mapping is performed\\
\verb"load_stab"& 0& 0 = generate a new sample\\
& & 1 = load a previously created sample \\
\verb"alpha2_stab"& 0.4& critical value for correlations $\rho$ in filtered samples:\\
\verb"pvalue_corr"& 0.001& critical p-value for correlations $\rho$ in filtered samples:\\
& & plot couples of parameters with \\
& & $|\rho|>$\verb"alpha2_stab"\\
\verb"ksstat"& 0.1& critical value for Smirnov statistics $d$: \\
& & plot parameters with $d>$\verb"ksstat"\\ \hline
& & p-value$<$\verb"pvalue_corr"\\
\verb"pvalue_ks" & 0.001& critical p-value for Smirnov statistics $d$: \\
& & plot parameters with p-value$<$\verb"pvalue_ks"\\
\verb"lik_init" & 1& 1 = the model is stationary (unit roots are `explosive')\\
& & 3 = the model has unit roots (unit roots are `stable')\\ \hline
\end{tabular}
\newpage
@ -167,10 +172,13 @@ out-of-sample validation. \vspace{0.5cm}
& & of reduced form coefficients\\
& & [\verb"max" \verb"max"] = analyse filtered \\
& & entries within the range [\verb"max" \verb"max"] \\
\verb"ksstat_redform"& 0.001& p-value for Smirnov statistics $d$ \\
& & when reduced form entries are filtered\\
\verb"alpha2_redform"& 0.001& p-value for correlation $\rho$ \\
& & when reduced form entries are filtered\\
\verb"ksstat_redform"& 0.001& critical p-value for Smirnov statistics $d$ \\
& & when \verb"threshold_redform" is active\\
& & plot parameters with p-value$<$\verb"ksstat_redform"\\
\verb"alpha2_redform"& 0.001& critical p-value for correlation $\rho$ \\
& & when \verb"threshold_redform" is active\\
& & plot couples of parameters with \\
& & p-value$<$\verb"alpha2_redform"\\
\verb"namendo"& () & list of endogenous variables \\
& : & jolly character to indicate ALL endogenous \\
\verb"namlagendo"& () & list of lagged endogenous variables:\\
@ -212,6 +220,7 @@ options of \verb"dynare_estimation". These are:
\item \verb"datafile"
\item \verb"mode_file"
\item \verb"first_obs"
\item \verb"lik_init"
\item \verb"nobs"
\item \verb"prefilter"
\item \verb"presample"
@ -235,10 +244,10 @@ options of \verb"dynare_estimation". These are:
\verb"istart_rmse"& 1& start computing RMSE's from \verb"istart_rmse":\\
& & use 2 to avoid big initial error \\
\verb"alpha_rmse"& 0.001& p-value for Smirnov statistics $d$:\\
& & plot parameters with $pvalue<$\verb"alpha_rmse"\\
& & plot parameters with p-value$<$\verb"alpha_rmse"\\
\verb"alpha2_rmse"& 0.001& p-value for correlation $\rho$\\
& & plot couples of parameters with
$pvalue<$\verb"alpha2_rmse"\\
p-value$<$\verb"alpha2_rmse"\\
\end{tabular}
\subsection{Screening analysis}
@ -248,14 +257,12 @@ with respect to those listed in the `Sampling options':
performs all the analyses required and displays results.
\subsection{Identification analysis (under development)}
\subsection{Identification analysis}
Setting the option \verb"identification=1", an identification
analysis based on theoretical moments is performed. Sensitivity plots are provided that
allow to infer which parameters are most likely to be less
identifiable.
Prerequisite for properly running all the identification routines, is the keyword
\verb"identification;" in the DYNARE model file. This keyword triggers the computation of analytic derivatives of the model with respect to estimated parameters and shocks. This is required for options \verb"morris=2" below, which implement \cite{Iskrev2009} identification analysis.
\vspace{1cm}
@ -263,36 +270,23 @@ Prerequisite for properly running all the identification routines, is the keywor
option name & default & description \\ \hline
\verb"identification"& 0 & 0 = no identification analysis \\
& & 1 = performs identification analysis:\\
& & this forces \verb"redform"=0 and default\verb"morris"=1\\
& & this forces \verb"redform"=0 and default \verb"morris"=1\\
\verb"morris"& 1 & 1 = Screening analysis (Type II error)\\
& & 0 = ANOVA Mapping (Type I error)\\
& & 2 = Analytic derivatives (similar to Type II error)\\
& & 2 = Analytic derivatives \citep{Iskrev2010,Iskrev2011}\\
\verb"morris_nliv"& 6 & number of levels in Morris design\\
\verb"morris_ntra"& 20& number of trajectories in Morris design\\
\verb"trans_ident"& 0 & data transformation for ANOVA Mapping\\
& & 0 = no transformation\\
& & 1 = log-transformation\\
& & 2 = rank-transformation\\\hline
\end{tabular}
\vspace{1cm}
\noindent For example, the following commands in the DYNARE model file
\vspace{1cm}
\noindent\verb"identification;"\\
\verb"dynare_sensitivity(identification=1, morris=2);"
\noindent\verb"dynare_sensitivity(identification=1, morris=2);"
\vspace{1cm}
\noindent trigger the identification analysis using analytic derivatives \citep{Iskrev2009}, jointly with the mapping of the acceptable region.
\noindent trigger the identification analysis using \cite{Iskrev2010,Iskrev2011}, jointly with the mapping of the acceptable region.
The identification analysis with derivatives can also be triggered by the commands
\vspace{1cm}
\noindent\verb"identification;"\\
\verb"dynare_identification;"
\vspace{1cm}
This does not do the mapping of acceptable regions for the model and uses the standard random sampler of DYNARE. It completely offsets any use of the sensitivity analysis toolbox.
\newpage
\section{Directory structure}
Sensitivity analysis results are saved on the hard-disk of the

24
doc/gsa/marco.bib
View File

@ -1,10 +1,3 @@
@TECHREPORT{Iskrev2009,
AUTHOR={Iskrev, Nikolay},
TITLE={Local Identification in \textsc{DSGE} Models},
YEAR=2009,
TYPE={mimeo},
}
@ARTICLE{Ratto_CompEcon_2008,
author = {Ratto, M.},
title = {Analysing DSGE Models with Global Sensitivity Analysis},
@ -13,3 +6,20 @@ TYPE={mimeo},
volume = {31},
pages = {115--139},
}
@ARTICLE{Iskrev2010,
author = {Nikolay Iskrev},
title = {Local Identification in {DSGE} Models},
journal = {Journal of Monetary Economics},
year = {2010},
volume = {57},
pages = {189-202}
}
@UNPUBLISHED{Iskrev2011,
author = {Nikolay Iskrev},
title = {Evaluating the strenght of identification in {DSGE} models. An a priori
approach},
note = {mimeo},
year = {2011}
}