Welcome to Francesca Molinari's homepage.
Working papers and Work in Progress
Published and accepted articles
STATA Software for Best Linear Prediction with Interval Outcome Data
Curriculum Vitae (in pdf format, updated January 2013)
Department of Economics Email: firstname.lastname@example.org Cornell University http://www.arts.cornell.edu/econ/fmolinari/ 458 Uris Hall Phone: 607-255-6367 Ithaca, NY 14853-7601, USA Fax: 607-255-2818
Working Papers and Work in Progress
Estimation and Inference for Best Linear Approximations to Set Identified Functions, with Arun G. Chandrasekhar, Victor Chernozhukov, and Paul Schrimpf. CeMMAP Working Papers # CWP43/12, December 2012.***
Unlucky or Risky? Unobserved Heterogeneity and Experience Rating in Insurance Markets, with Levon Barseghyan, Darcy Steeg Morris and Josh Teitelbaum,November 2012. Submitted.***,****
Inference Under Stability of Risk Preferences, with Levon Barseghyan, and Josh Teitelbaum. Coming soon.****
Computational Methods for Partially Identified Models via Data Augmentation and Support Vector Machines, with Haim Bar . Coming soon.***
Bracketing and Risk Preferences: Identification and Estimation in Field Data, with Levon Barseghyan, Ted O'Donoghue and Josh Teitelbaum. Coming soon.****
Published and forthcoming articles
Applications of Random Set Theory in Econometrics, with Ilya Molchanov, July 2013. Forthcoming in the Annual Review of Economics.***
The Nature of Risk Preferences: Evidence from Insurance Choice, with Levon Barseghyan, Ted O'Donoghue and Josh Teitelbaum, November 2012. Forthcoming in the American Economic Review.*,***,****
Distinguishing Probability Weighting from Risk Misperceptions in Field Data, with Levon Barseghyan, Ted O'Donoghue and Josh Teitelbaum, American Economic Review Papers and Proceedings, 2013, vol. 103, number 3, 580-585.***,****
Partial Identification Using Random Set Theory, with Arie Beresteanu and Ilya Molchanov, Journal of Econometrics, 2012, vol. 166, issue 1, pages 17-32. ERRATA.*,***
Sharp Identification Regions in Models with Convex Moment Predictions, with Arie Beresteanu and Ilya Molchanov, Econometrica, 2011, vol. 79, issue 6, pages 1785-1821; with Online Supplement.*,*** (This paper supercedes three earlier working papers by the same authors: CeMMAP Working Papers # CWP25/10, CWP27/09 and CWP15/08.)
Rounding Probabilistic Expectations in Surveys, with Chuck Manski, Journal of Business and Economic Statistics, 2010, vol. 28, number 2, pages 219-231.*,**
Missing Treatments, Journal of Business and Economic Statistics, 2010, vol. 28, number 1, pages 82-95.*
The Identification Power of Equilibrium in Games: The Supermodular Case, with Adam Rosen, Journal of Business and Economic Statistics, 2008, vol. 26, number 3, pages 297-302. Invited discussion of Aradillas-Lopez and Tamer (2008) prepared for the 2007 Joint Statistical Meetings.*
Asymptotic Properties for a Class of Partially Identified Models, with Arie Beresteanu, Econometrica, 2008, vol. 76, issue 4, pages 763-814.*
Partial Identification of Probability Distributions with Misclassified Data, Journal of Econometrics, 2008, vol. 144, issue 1, pages 81-117.*
Skip Sequencing: A Decision Problem in Questionnaire Design, with Chuck Manski, Annals of Applied Statistics, 2008, vol. 2, number 1, 264-285.*,**
Spatial Correlation Robust Inference with Errors in Location and Distance, with Timothy G. Conley, Journal of Econometrics, 2007, vol. 140, issue 1, pages 76-96. Download the Technical Appendix here.
Generalization of a Result on 'Regressions: Short and Long', with Marcin Peski, Econometric Theory, 2006, vol. 22, issue 01, pages 159-163.
STATA Software for Best Linear Prediction with Interval Outcome Data*
Click here to download a zip file containing STATA .ado files and .sthlp files to replicate the results in Beresteanu and Molinari (2008) and implement their theoretical results. The code allows for best linear prediction with any number of perfectly observed covariates. It returns estimates of the sharp identification region for each component of the BLP parameter vector, as well as confidence sets and confidence collections for each of these regions. It also allows for hypothesis testing and confidence statements on any pair of (functions of) elements of the BLP parameter vector. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of merchantability or fitness for a particular purpose. If you use the software, we ask that you please cite it as: Arie Beresteanu, Francesca Molinari and Darcy Steeg Morris (2010): "Asymptotics for Partially Identified Models in STATA," and that you also cite Beresteanu and Molinari (Econometrica, 2008) as the source of the theoretical results.
* This research was in part supported by the National Science Foundation grant SES-0617482.
** This research was in part supported by the National Institute of Aging grant R21 AG028465-01.
*** This research was in part supported by the National Science Foundation grant SES-0922330.
**** This research was in part supported by the National Science Foundation grant SES-1031136.
Last revised: January 2013.