gragusa.org on gragusa.org
https://gragusa.org/index.xml
en-usGiuseppe RagusaCopyright (c) Giuseppe Ragusa.Fri, 01 Dec 2017 00:00:00 UTCBayesian Estimation of State Space Models Using Moment Conditions
https://gragusa.org/publications/bayesian-state-space/
Fri, 01 Dec 2017 00:00:00 UTCGiuseppe Ragusahttps://gragusa.org/publications/bayesian-state-space/<p>We consider Bayesian estimation of state space models when the measurement density is not available but estimating equations for the parameters of the measurement density are available from moment conditions. The most common applications are partial equilibrium models involving moment conditions that depend on dynamic latent variables (e.g., timevarying parameters, stochastic volatility) and dynamic general equilibrium models when moment equations from the first order conditions are available but computing an accurate approximation to the measurement density is difficult.</p>
Anchoring the Yield Curve Using Survey Expectations
https://gragusa.org/publications/anchoring/
Fri, 01 Sep 2017 00:00:00 UTCGiuseppe Ragusahttps://gragusa.org/publications/anchoring/<p>The dynamic behavior of the term structure of interest rates is difficult to replicate with mod- els, and even models with a proven track record of empirical performance have underperformed since the early 2000s. On the other hand, survey expectations can accurately predict yields, but they are typically not available for all maturities and/or forecast horizons. We show how survey expectations can be exploited to improve the accuracy of yield curve forecasts given by a base model. We do so by employing a flexible exponential tilting method that anchors the model forecasts to the survey expectations, and we develop a test to guide the choice of the anchoring points. The method implicitly incorporates into yield curve forecasts any information that survey participants have access to - such as information about the current state of the economy or forward-looking information contained in monetary policy announcements - without the need to explicitly model it. We document that anchoring delivers large and significant gains in forecast accuracy relative to the class of models that are widely adopted by financial and policy institutions for forecasting the term structure of interest rates.</p>
Sensitivity, Moment Conditions, and the Risk-free Rate in Yogo (2006)
https://gragusa.org/publications/moment-conditions-sensitivity/
Fri, 01 Sep 2017 00:00:00 UTCGiuseppe Ragusahttps://gragusa.org/publications/moment-conditions-sensitivity/<p>In this paper we show that results presented in the seminal paper by Yogo, A Consumption Based Explanation of Expected Stock Returns, cannot be replicated. We find different estimates for the parameters and we obtain values of over-identified statistics that being much larger than those in the original paper indicate rejection of the durable consumption asset pricing model. By careful inspection of Yogo’s replication files, we were able to track down the inconsistency to a coding bug. The rejection of the durable model is exemplified by its inability to simultaneously explain the risk-free rate and excess stock returns.</p>
From Empty Pews to Empty Cradles: Fertility Decline Among European Catholics
https://gragusa.org/publications/empypub/
Thu, 01 Jun 2017 00:00:00 UTCGiuseppe Ragusahttps://gragusa.org/publications/empypub/<p>Catholic countries of Europe pose a demographic puzzle –fertility is unprecedentedly low (total fertility$=$1.3) despite low female labor force participation. We model three channels of religious effects on demand for children: through changing norms, reduced market wages, and reduced costs of childrearing. We estimate their effects using new panel data on church attendance and clergy employment for thirteen European countries from 1960-2000, spanning the Second Vatican Council (1962-65). Catholic theology is uniform across countries. Yet service varied considerably across countries and over time, especially before the Council, reflecting differences in Church provision of education, health, welfare and other social services. We use differential declines in service provision –measured by nuns/capita– to identify its effect on fertility, controlling for secular trends. They are large: 300 to 400 children per nun. Reduced religiosity (measured by church attendance) has no effect for Protestants, but predicts fertility decline for Catholics. The data suggest that service provision and religiosity complement each other –a finding consistent with preferential provision of services to church attendees. Nuns outperform priests in predicting fertility, suggesting that the childrearing cost channel dominates theology and norms.</p>
Econometrics of DSGE models
https://gragusa.org/teaching/eief-dsge/
Thu, 16 Feb 2017 00:00:00 +1100Giuseppe Ragusahttps://gragusa.org/teaching/eief-dsge/<p>This is a course on the econometric techniques used in the estimation of dynamic
macroeconomic models (DSGE models). The aim of the course is mostly theoretical, but applications are also presented using <a href="http://julialang.org">Julia</a>.</p>
<p></p>
<h3 id="topics">Topics</h3>
<ol>
<li>Motivation: DSGE models and their applications</li>
<li><p>Approximating and solving DSGE models</p>
<p>a. State space representation
b. Constructing (log-)linear approximation</p></li>
<li><p>Time series properties of the model and data</p></li>
<li><p>Classical estimation of DSGE models</p>
<p>a. Generalized Method of Moments (GMM)</p>
<p>b. Simulated Method of Moments (SMM) and Indirect Inference (IF)</p>
<p>c. Impulse response functions matching</p></li>
<li><p>Bayesian estimation of DSGE models</p>
<p>a. (log-)linear models</p>
<p>b. Non linear models</p></li>
<li><p>The twilight zone of DSGE estimation</p>
<p>a. Identification</p>
<p>b. Feasible non linear estimation</p>
<p>c. VAR and DSGE</p>
<p>d. Limited information estimation</p></li>
</ol>
<h3 id="readings">Readings</h3>
<p>This is a list of readings. This is by no means comprehensive — it
simply reflects the source that are closely related with the topics
covered in lectures.</p>
<h4 id="books">Books:</h4>
<ul>
<li>Canova, F. (2007), Methods for Applied Macroeconomic Research,
Princeton: Princeton University Press.</li>
<li>Dejong, D.N. and C. Dave (2007), Structural Macroeconomics,
Princeton: Princeton University Press</li>
</ul>
<h4 id="papers">Papers:</h4>
<ul>
<li>An and Schoerfheide (2007) Bayesian Analysis of DSGE Models,
Econometric Reviews, 26(2-4), 2007, 113-172.
<a href="http://www.tandfonline.com/doi/abs/10.1080/07474930701220071">Download</a></li>
<li>Fernández-Villaverde, Jesús. “The econometrics of DSGE models.”
SERIEs 1.1-2 (2010): 3-49.
<a href="http://link.springer.com/article/10.1007/s13209-009-0014-7">Download</a></li>
<li>Schorfheide, Frank. “Loss function-based evaluation of DSGE models.”
Journal of Applied Econometrics 15.6 (2000): 645-670.
<a href="http://onlinelibrary.wiley.com/doi/10.1002/jae.582/full">Download</a></li>
<li>Arulampalam, M. Sanjeev, et al. “A tutorial on particle filters for
online nonlinear/non-Gaussian Bayesian tracking.” Signal Processing,
IEEE Transactions on 50.2 (2002): 174-188.
<a href="http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=978374">Download</a></li>
<li>Doucet, Arnaud, and Adam M. Johansen. “A tutorial on particle
filtering and smoothing: Fifteen years later.” Handbook of Nonlinear
Filtering 12.656-704 (2009): 3.
<a href="http://automatica.dei.unipd.it/tl_files/utenti/lucaschenato/Classes/PSC10_11/Tutorial_PF_doucet_johansen.pdf">Download</a></li>
<li>Andrieu, Christophe, Arnaud Doucet, and Roman Holenstein. “Particle
markov chain monte carlo methods.” Journal of the Royal Statistical
Society: Series B (Statistical Methodology) 72.3 (2010): 269-342.
<a href="http://onlinelibrary.wiley.com/doi/10.1111/j.1467-9868.2009.00736.x/full">Download</a></li>
<li>Gallant, A. Ronald, Raffaella Giacomini, and Giuseppe Ragusa. Bayesian
estimation of state space models using moment conditions. Technical
report, 2015. <a href="http://www.aronaldg.org/papers/bliml.pdf">Download</a></li>
<li>Giacomini, Raffaella. “The relationship between DSGE and VAR models.”
Advances in Econometrics 31 (2013).
<a href="http://www.emeraldinsight.com/doi/abs/10.1108/S0731-9053(2013)0000031001">Download</a></li>
<li>Komunjer, Ivana, and Serena Ng. “Dynamic identification of
stochastic general equilibrium models.” Econometrica 79.6
(2011):1995-2032.
<a href="http://www.columbia.edu/~sn2294/pub/ecta11.pdf">Download</a></li>
<li>Del Negro, Marco, and Frank Schorfheide. “Bayesian
macroeconometrics.” The Oxford handbook of Bayesian econometrics 293
(2011): 389.
<a href="http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.414.4871&rep=rep1&type=pdf">Download</a></li>
</ul>
<h3 id="julia">Julia</h3>
<p><a href="http://julialang.org">Julia</a> is a high-level, high-performance dynamic
programming language for technical computing, with syntax that is
familiar to users of other technical computing environments. In
particular, its syntax is similar to Matlab. The similarity of the
syntax means that a lot of Matlab code will run almost unmodified.</p>
<p><code>Julia</code> has many advantages over other languages and for this reason is
being extensively used in industries and in research.</p>
<p>Recently, the <a href="https://www.newyorkfed.org/">Federal Reserve of New York</a> has
open sourced its macroeconomic model (used for producing forecast about key
variables and to conduct policy experiment). The code is written in Julia. You
can read about
it
<a href="http://libertystreeteconomics.newyorkfed.org/2015/12/the-frbny-dsge-model-meets-julia.html">here</a>.
The code is <a href="https://github.com/FRBNY-DSGE/DSGE.jl">here</a>.</p>
<p><a href="http://quant-econ.net/jl/learning_julia.html">Programming in Julia</a> in an
excellent tutorial is written by Thomas J. Sargent and John Stachurski. Along
with being a very good introduction to the language, this is also a complete macroeconomic textbook with concept illustrated in Julia.
.</p>
<h2 id="slides">slides</h2>
<ul>
<li><a href="files/teaching/lecture_1.pdf">Lecture 1</a></li>
<li><a href="files/teaching/lecture_2.pdf">Lecture 2</a></li>
<li><a href="files/teaching/lecture_4.pdf">Lecture 4</a></li>
<li><a href="files/teaching/lecture_5.pdf">Lecture 5</a> <a href="files/teaching/MH.ipynb">Jupyter notebook</a> <a href="files/teaching/PS-1.pdf">Problem set #1</a></li>
<li><a href="files/teaching/lecture_6.pdf">Lecture 6</a></li>
<li><a href="files/teaching/lecture_7.pdf">Lecture 7</a></li>
<li><a href="files/teaching/lecture_9.pdf">Lecture 9</a> <a href="files/teaching/particle_filter.ipynb">Jupyter notebook</a></li>
</ul>Applied Econometrics and Statistics
https://gragusa.org/teaching/ase/
Mon, 13 Feb 2017 00:00:00 +1100Giuseppe Ragusahttps://gragusa.org/teaching/ase/<p>The aim of this course is to provide an introduction to the practice of
econometrics. While both theoretical and practical aspects will be covered,
emphasis will be on intuitive understanding. Concepts will be illustrated with
real world applications on real world data.</p>
<p>
<span class="label error outline"><a href="../files/teaching/syllabus-ase-2017.pdf">Syllabus</a></span>
<span class="label outline"><a href="http://gragusa.org/ase">Class website</a></span></p>Econometric Theory
https://gragusa.org/teaching/grad-et/
Mon, 13 Feb 2017 00:00:00 +1100Giuseppe Ragusahttps://gragusa.org/teaching/grad-et/<p>The course main objective is to provide students with a solid grasp of the
probabilistic and statistical underpinnings of the most common econometric
techniques. The topics studied will be of practical use to any student who plans
on confronting data in their thesis or wishes to read and precisely understand
the econometrics typically used in empirical research published in academic
journals. The recommended software package is <a href="http://www.julialang.org">Julia</a>.</p>
<p></p>
<p>Econometric theory has two parts. The first part focuses on the specification
and estimation of the linear regression models and its extensions. The second
part focuses on the theory and the applications of time series methods in
econometrics.</p>
<p><span class="label outline"><a href="http://gragusa.org/et">Class website</a></span></p>Problem sets
https://gragusa.org/ase/problemsets/
Sat, 11 Feb 2017 00:00:00 UTCGiuseppe Ragusahttps://gragusa.org/ase/problemsets/
<p>There will be a weekly problem set. The assignments will involve both theoretical and empirical work. Group study and free discussion are encouraged. But you should submit your own answers. You will probably find the class very hard to follow if you fail to spend sufficient time on all of the problem sets. The problem sets are part of the final grade as explained in the next section below.</p>
<p>If you have any question on the problem sets, please ask me or TA’s during our office hours. Our office hours are for you. I prefer to talk to you in person. I feel that Email is not a very efficient way to ask econometric questions.</p>
<p>Problem set answers are to be turned in on time. You can hand in the homework AFTER the class. Please do not come and hand it in to me whilst I am lecturing. Do not email assignments. Late solution will not be accepted!</p>
<h4 id="problem-sets">Problem sets</h4>
<ul>
<li><p><a href="http://docenti.luiss.it/protected-uploads/580/2017/02/20170215192957-P_1.pdf">Problem Set 1</a></p></li>
<li><p>Problem Set 2 (Due Tuesday, March, 7 2017)</p>
<ul>
<li><a href="http://docenti.luiss.it/protected-uploads/580/2017/03/20170302161751-PS02_Questions.pdf">Questions</a></li>
<li><a href="http://docenti.luiss.it/protected-uploads/580/2017/03/20170302161740-PS02_Answers.pdf">Answer sheet</a></li>
</ul>
<p><em>Note:</em> This problem set is multiple choice. You have to print the <a href="http://docenti.luiss.it/protected-uploads/580/2017/03/20170302161740-PS02_Answers.pdf">answer sheet</a>, fill your id, and fill your answers based on the <a href="http://docenti.luiss.it/protected-uploads/580/2017/03/20170302161751-PS02_Questions.pdf">questions</a>.</p></li>
<li><p>Problem Set 3 (Due Thursday, March 16 2017)
This problem set is personalized. Each student must download her own version of the problem set. To download the problem set use the following URL:
<code>http://gragusa.org/ps3/xxxxxx.pdf</code> where <code>xxxxxx</code> is the LUISS ID of the student.</p></li>
<li><p>Problem Set 4 (Due Tuesday, April 11 2017)
This problem set is personalized. Each student must download her own version of the problem set. To download the problem set use the following URL:
<code>http://gragusa.org/ps4/xxxxxx.pdf</code> where <code>xxxxxx</code> is the LUISS ID of the student.</p></li>
<li><p>Problem Set 5 (Due Tuesday, May 2 2017)
This problem set is personalized. Each student must download her own version of the problem set. To download the problem set use the following URL:
<code>http://gragusa.org/ps5/xxxxxx.pdf</code> where <code>xxxxxx</code> is the LUISS ID of the student.</p></li>
<li><p><a href="../files/teaching/PS_06.pdf">Problem Set 6</a></p></li>
</ul>
tasessions et
https://gragusa.org/et/tasessions/
Sat, 11 Feb 2017 00:00:00 UTCGiuseppe Ragusahttps://gragusa.org/et/tasessions/<p>The TAs will lead a weekly practice session which will be held in the computer labs (301 and 306).</p>
<p>The sessions are an important part of the course and regular and
attendance is strongly advised. During these sessions, the TA will review the
concepts introduced in class and solve applied problems using Julia. Sessions are
also useful for asking questions about the problem sets.</p>
<p>The following table gives the distribution of students across the two sessions:</p>
<table>
<thead>
<tr>
<th align="left">TA</th>
<th align="left">Time</th>
<th align="left">Students</th>
</tr>
</thead>
<tbody>
<tr>
<td align="left">Siria Angino</td>
<td align="left">10:15-11:45</td>
<td align="left">J-Z</td>
</tr>
<tr>
<td align="left">Chiara Perricone</td>
<td align="left">8:45-10:25</td>
<td align="left">A-I</td>
</tr>
</tbody>
</table>
Problem sets et
https://gragusa.org/et/problemsets/
Fri, 10 Feb 2017 00:00:00 UTCGiuseppe Ragusahttps://gragusa.org/et/problemsets/
<p>There will be a weekly problem set. The assignments will involve both theoretical and empirical work. Group study and free discussion are encouraged. But you should submit your own answers. You will probably find the class very hard to follow if you fail to spend sufficient time on all of the problem sets. The problem sets are part of the final grade as explained in the next section below.</p>
<p>If you have any question on the problem sets, please ask me or TA’s during our office hours. Our office hours are for you. I prefer to talk to you in person. I feel that email is not a very efficient way to ask econometric questions.</p>
<p>Problem set answers are to be turned in on time. You can hand in the homework AFTER the class. Late solution will not be accepted!</p>
<h4 id="problem-set">Problem set</h4>
<ul>
<li><p><a href="http://docenti.luiss.it/perricone/files/2017/02/PS_01.pdf">Problem Set 1</a></p></li>
<li><p><a href="http://docenti.luiss.it/perricone/files/2017/02/PS_02.pdf">Problem Set 2</a></p></li>
<li><p><a href="http://docenti.luiss.it/perricone/files/2017/02/PS_03.pdf">Problem Set 3</a></p></li>
<li><p><a href="http://docenti.luiss.it/perricone/files/2017/02/PS_04.pdf">Problem Set 4</a></p></li>
<li><p><a href="http://docenti.luiss.it/perricone/files/2017/02/PS_05.pdf">Problem Set 5</a></p></li>
</ul>