All seminars are held in 1011 Evans Hall at UC Berkeley, unless otherwise notified.

Upcoming seminar

Tuesday, January 22, 2019 11:00 AM to 12:30 PM

Peng Ding, UC Berkeley: Instrumental variables as bias amplifiers with general outcome and confounding

Drawing causal inference with observational studies is the central pillar of many disciplines. One sufficient condition for identifying the causal effect is that the treatment-outcome relationship is unconfounded conditional on the observed covariates. It is often believed that the more covariates we condition on, the more plausible this unconfoundedness assumption is. This belief has had a huge impact on practical causal inference, suggesting that we should adjust for all pretreatment covariates. However, when there is unmeasured confounding between the treatment and outcome, estimators adjusting for some pretreatment covariate might have greater bias than estimators that do not adjust for this covariate. This kind of covariate is called a bias amplifier and includes instrumental variables that are independent of the confounder and affect the outcome only through the treatment. Previously, theoretical results for this phenomenon have been established only for linear models. We fill this gap in the literature by providing a general theory, showing that this phenomenon happens under a wide class of models satisfying certain monotonicity assumptions.

All seminars


Tuesday, January 22, 2019 11:00 AM to 12:30 PM
Peng Ding, UC Berkeley: Instrumental variables as bias amplifiers with general outcome and confounding

Drawing causal inference with observational studies is the central pillar of many disciplines. One sufficient condition for identifying the causal effect is that the treatment-outcome relationship is unconfounded conditional on the observed covariates. It is often believed that the more covariates we condition on, the more plausible this unconfoundedness assumption is. This belief has had a huge impact on practical causal inference, suggesting that we should adjust for all pretreatment covariates. However, when there is unmeasured confounding between the treatment and outcome, estimators adjusting for some pretreatment covariate might have greater bias than estimators that do not adjust for this covariate. This kind of covariate is called a bias amplifier and includes instrumental variables that are independent of the confounder and affect the outcome only through the treatment. Previously, theoretical results for this phenomenon have been established only for linear models. We fill this gap in the literature by providing a general theory, showing that this phenomenon happens under a wide class of models satisfying certain monotonicity assumptions.



Tuesday, January 29, 2019 11:00 AM to 12:30 PM
Matteo Basei, UC Berkeley: The coordination of centralised and distributed generation
We analyse the interaction between centralised carbon-emissive technologies and distributed non-emissive technologies. A representative consumer can satisfy her electricity demand by investing in solar panels and by buying power from a centralised firm. We consider the point of view of the consumer, the firm and a social planner, formulating suitable McKean-Vlasov control problems with stochastic coefficients. First, we provide explicit formulas for the production strategies which minimise the costs. Then, we look for an equilibrium price. Joint work with René Aid and Huyen Pham.


Tuesday, February 5, 2019 11:00 AM to 12:30 PM


Tuesday, February 12, 2019 11:00 AM to 12:30 PM


Tuesday, February 19, 2019 11:00 AM to 12:30 PM


Tuesday, February 26, 2019 11:00 AM to 12:30 PM


Tuesday, March 5, 2019 11:00 AM to 12:30 PM


Tuesday, March 12, 2019 11:00 AM to 12:30 PM


Tuesday, March 19, 2019 11:00 AM to 12:30 PM


Tuesday, April 2, 2019 11:00 AM to 12:30 PM


Tuesday, April 9, 2019 11:00 AM to 12:30 PM


Tuesday, April 16, 2019 11:00 AM to 12:30 PM


Tuesday, April 23, 2019 11:00 AM to 12:30 PM


Tuesday, April 30, 2019 11:00 AM to 12:30 PM