HOME /

Presentations

4/25/18 - Jeff Gill presents "Models for Identifying Substantive Clusters and Fitted Subclusters in Social Science Data", at CGIS Knafel K354, Wednesday, April 25, 2018

Jeff Gill presents

 

Models for Identifying Substantive Clusters and Fitted Subclusters in Social Science Data

Unseen grouping, often called latent clustering, is a common feature in social science data.  Subjects may intentionally or unintentionially group themselves in ways that complicate the statistical analysis of substantively important relationships. This work introduces a new model-based clustering design which incorporates two sources of heterogeneity.  The first source is a random effect that introduces substantively unimportant...

Read more about 4/25/18 - Jeff Gill presents "Models for Identifying Substantive Clusters and Fitted Subclusters in Social Science Data"
4/18/18- Xiang Zhou presents "Two residual-based methods to adjust for treatment-induced confounding in causal inference", at CGIS Knafel K354, Wednesday, April 18, 2018

Xiang Zhou presents "Two residual-based methods to adjust for treatment-induced confounding in causal inference"

 

Two residual-based methods to adjust for treatment-induced confounding in causal inference 

 

Treatment-induced confounding arises in both causal inference of time-varying treatments and causal mediation analysis where post-treatment variables affect both the mediator and outcome. Existing methods to adjust for treatment-induced confounding include, among others, Robins's structural nest mean model (SNMM) with its g-... Read more about 4/18/18- Xiang Zhou presents "Two residual-based methods to adjust for treatment-induced confounding in causal inference"
4/11/18 - Michael Windzio presents "Does schoolwork cooperation improve pupils’ grades and well-being in school? Results from social network and propensity score analysis", at CGIS Knafel K354, Wednesday, April 11, 2018


Does schoolwork cooperation improve pupils’ grades and well-being in school? Results from social network and propensity score analysis


Using panel data of school-class networks and outcomes of 11-13-year-old students, effects of collaboration in schoolwork networks on grades and school-related well-being will be investigated. The analysis might suffer from endogeneity-bias because pupils actively select their peers also with regard to their school-performance. This selectivity will be demonstrated by using p* models for ties in schoolwork-networks at t1 based data of 1,289...

Read more about 4/11/18 - Michael Windzio presents "Does schoolwork cooperation improve pupils’ grades and well-being in school? Results from social network and propensity score analysis"
3/28/18 - Michelle Torres Pacheco presents "Understanding visual messages: visual framing and the Bag of Visual Words", at CGIS Knafel K354, Wednesday, March 28, 2018

Michelle Torres Pacheco presents. 

 

TITLE: Understanding visual messages: visual framing and the Bag of Visual Words

ABSTRACT
Political communication is a central element of several political dynamics. Its visual component is crucial in understanding the origin, characteristics and consequences of the messages sent between political figures, media and citizens. However, visual features have been largely overlooked in Political Science. In this project, I implement computer vision and image retrieval techniques to measure and understand messages...

Read more about 3/28/18 - Michelle Torres Pacheco presents "Understanding visual messages: visual framing and the Bag of Visual Words"
3/7/18 - Tianxiao Shen presents, "Language Style Transfer", at CGIS Knafel K354, Wednesday, March 7, 2018

Tianxiao Shen presents, "Language Style Transfer"

 

Abstract: Recent advances in text generation tasks such as machine translation and summarization rely on the use of massive amounts of parallel data, which is costly to collect or nonexistent in many scenarios. In this talk, I will present a novel model to perform style transfer on the basis of non-parallel text. This is an instance of a broad family of problems including machine translation, decipherment, and sentiment modification. I will talk about how we deal with the challenge to...

Read more about 3/7/18 - Tianxiao Shen presents, "Language Style Transfer"
2/28/18 - Isaac Kohane presents “Interesting early results in data science that require new statistical methods” Wednesday, February 28, 2018

Isaac Kohane presents “Interesting early results in data science that require new statistical methods”

 

Title: “Interesting early results in data science that require new statistical methods”

 

Abstract: In the brave new world of biomedical data science, new sources of data emerge seemingly every year from Twitter to genomes to weather to drug habits and/or doctor preferences.  I will outline several interesting and apparently impactful findings that have emerged as a result of analyses of these data, both individually and...

Read more about 2/28/18 - Isaac Kohane presents “Interesting early results in data science that require new statistical methods”
2/21/18 Emily Breza presents "Using Aggregated Relational Data to Feasibly Identify Network Structure without Network Data with Arun Chandrasekhar, Tyler McCormick and Mengjie Pan", at CGIS Knafel K354, Wednesday, February 21, 2018

Emily Breza presents "Using Aggregated Relational Data to Feasibly Identify Network Structure without Network Data with Arun Chandrasekhar, Tyler McCormick and Mengjie Pan"

 

Using Aggregated Relational Data to Feasibly Identify Network Structure without Network Data with Arun Chandrasekhar, Tyler McCormick and Mengjie Pan

 

Abstract. Social network data is often prohibitively expensive to collect, limiting empirical network research. Typical economic network mapping requires (1) enumerating a census, (2) eliciting the names of all...

Read more about 2/21/18 Emily Breza presents "Using Aggregated Relational Data to Feasibly Identify Network Structure without Network Data with Arun Chandrasekhar, Tyler McCormick and Mengjie Pan"

Pages