Presentations

2/15/17 - Dean Eckles (MIT Sloan) - "Estimating peer effects in networks with peer encouragement designs", at CGIS Knafel K354, Wednesday, February 15, 2017

Dean Eckles (MIT Sloan) presents "Estimating peer effects in networks with peer encouragement designs"

 

Peer effects, in which the behavior of an individual is affected by the behavior of their peers, are central to social science. Because peer effects are often confounded with homophily and common external causes, recent work has used randomized experiments to estimate effects of specific peer behaviors. These experiments have often relied on the experimenter being able to randomly modulate mechanisms by which peer behavior is transmitted to a focal individual....

Read more about 2/15/17 - Dean Eckles (MIT Sloan) - "Estimating peer effects in networks with peer encouragement designs"
2/8/17 - Michael Peress - Does Newspaper Coverage Mediate the Economic Vote?, at CGIS Knafel K354, Wednesday, February 8, 2017

Michael Peress presents "Does Newspaper Coverage Mediate the Economic Vote?"

 

Do voters punish incumbent governments when economic performance is poor or when the media report that economic performance is poor? We draw on an original data set of over 2 million newspaper articles reporting on the economy in 32 newspapers in 16 developed democracies over 32 years. We develop procedures for coding newspaper sentiment on the economy and apply our results to study the role of the media in driving the economic vote. Our results indicate voters respond directly to unemployment,...

Read more about 2/8/17 - Michael Peress - Does Newspaper Coverage Mediate the Economic Vote?
2/1/17 - Paul von Hippel (UT-Austin) - Maximum likelihood multiple imputation: A more efficient approach to repairing and analyzing incomplete data, at CGIS Knafel K354, Wednesday, February 1, 2017

Paul von Hippel (UT- Austin) presents 

 

Title: Maximum likelihood multiple imputation: A more efficient approach to repairing and analyzing incomplete data

 

Abstract: Maximum likelihood multiple imputation (MLMI) is a form of multiple imputation (MI) that imputes values conditionally on a maximum likelihood estimate of the parameters. MLMI contrasts with the most popular form of MI, posterior draw multiple imputation (PDMI), which imputes values conditionally on an estimate drawn at random from the posterior distribution of the parameters. Despite...

Read more about 2/1/17 - Paul von Hippel (UT-Austin) - Maximum likelihood multiple imputation: A more efficient approach to repairing and analyzing incomplete data
1/25/17 - Matt Taddy (Chicago Booth), at CGIS Knafel K354, Wednesday, January 25, 2017

Matt Taddy (Chicago Booth) presentation 

 

Title:  Counterfactual Prediction with Deep Instrumental Variables Networks

 

Abstract:

We are in the middle of a remarkable rise in the use and capability of artificial intelligence. Much of this growth has been fueled by the success of deep learning architectures: models that map from observables to outputs via multiple layers of latent representations. These deep learning algorithms are effective tools for unstructured prediction, and they can be combined in AI systems to solve complex automated...

Read more about 1/25/17 - Matt Taddy (Chicago Booth)

Pages