Presentations

Growing a Pattern From a Seed- Presenter: Cynthia Rudin Wednesday, October 9, 2013

Presenter: Cynthia Rudin 

Abstract: I will describe two methods and applications for pattern detection, where patterns are grown from a seed of a few items:

1) Growing a List: The next generation of search engines should not simply retrieve URLs, but should aim at retrieving information. We designed a system that leads into this next generation, leveraging information from across the Internet to grow an authoritative list on almost any topic, starting from a seed.

2) Crime Series Detection: In joint work with the...

Read more about Growing a Pattern From a Seed- Presenter: Cynthia Rudin
Between the ordinal and the interval: Educational score scales and the "scale-dependence" of analytic results- Presenter: Andrew Ho Wednesday, October 2, 2013

Presenter: Andrew Ho 

Abstract: At some point early in our quantitative training, we learn the difference between ordinal scales and interval scales. There is a clear distinction. Ordinal scales, such as rankings, contain only information about order (1st place, 2nd place, 3rd place). Interval scales, such as time and temperature, have equal intervals between successive scale points (1 degree Celsius, 2 degrees Celsius, 3 degrees Celsius). We learn that these differences matter, because the vast majority of statistical analyses, from...

Read more about Between the ordinal and the interval: Educational score scales and the "scale-dependence" of analytic results- Presenter: Andrew Ho
Deterministic and Stochastic Counterfactuals, Interference Between Treatments, Causal Interactions, Bell's Inequality in Quantum Mechanics, and The Nature of Reality- Presenter: James Robins Wednesday, September 25, 2013

Presenter: James Robins 

Abstract: Neyman introduced a formal mathematical theory of counterfactual causation that now has become standard language in many quantitative disciplines, but not in physics. We use results on causal interaction and interference between treatments (derived under the Neyman theory) to give a simple new proof of a well-known result in quantum physics, namely, Bell's inequality.

Now the predictions of quantum mechanics and the results of experiment both violate Bell's inequality....

Read more about Deterministic and Stochastic Counterfactuals, Interference Between Treatments, Causal Interactions, Bell's Inequality in Quantum Mechanics, and The Nature of Reality- Presenter: James Robins
Propensity score weighting for covariate balance- Presenter: Alan M. Zaslavsky Wednesday, September 18, 2013

Presenter: Alan M. Zaslavsky 

Abstract: Balance of covariate distributions is crucial for an unconfounded descriptive or causal comparison between different groups. However, lack of overlap in the covariates is common in observational studies. We discuss weighting strategies for balancing covariates. A general class of weights --- the balancing weights --- that balance the expectation of the covariates in the treatment and the control groups is proposed. The methods rely on the propensity score and include several existing weights,...

Read more about Propensity score weighting for covariate balance- Presenter: Alan M. Zaslavsky
Does Regression Produce Representative Estimates of Causal Effects? Wednesday, September 11, 2013

Presenter: Peter Aronow 

Abstract: It is well-known that, with an unrepresentative sample, the estimate
of a causal effect may fail to characterize how effects operate in the
population of interest. What is less well understood is that
conventional estimation practices for observational studies may
produce the same problem even with a representative sample.
Specifically, causal effects estimated via multiple regression
differentially weight each unit's contribution. The ``effective
sample'' that regression uses to...

Read more about Does Regression Produce Representative Estimates of Causal Effects?
Discovering latent network structure from spiking data using Hawkes process Wednesday, September 4, 2013

Presenter: Scott W. Linderman 

Abstract: Scott Linderman will be speaking on discovering latent network structure from spiking data using Hawkes processes. Examples of this include: recovering neural connectivity from electrophysiological recordings; discovering interactions between stocks on the S&P 100; and identifying patterns in homicide data from Chicago.

Only Audio

...
Read more about Discovering latent network structure from spiking data using Hawkes process

Pages