## Date and Location

Feb 25, 2019 - 1:00pm to 2:00pm## Abstract

## Estimating Causal Effect from Observations (presented by Chandana Upadhyaya, Computer Science)

Cosma Rohilla Shalizi. (2019). Estimating Causal Effect from Observations. *Advanced Data Analysis from an Elementary Point of View*. (Chapter 23, pp 513 - 524).

Chapter 22 [presented on Feb 11] gave us ways of identifying causal effects, that is, of knowing when quantities like Pr (Y = y|do(X = x)) are functions of the distribution of observable variables. Once we know that something is identifiable, the next question is how we can actually estimate it from data.

## Discovering Causal Structure from Observations (presented by Nidhi Hiremath, Computer Science)

Cosma Rohilla Shalizi (2019) Discovering Causal Structure from Observations. *Advanced Data Analysis from an Elementary Point of View*. (Chapter 24, pp 525 - 542).

The last few chapters [presentations] have, hopefully, convinced you that when you want to do causal inference, it would help to know the causal graph. We have seen how the graph would let us calculate the effects of actual or hypothetical manipulations of the variables in the system. Furthermore, the graph tells us about what effects we can and cannot identify, and estimate, from observational data. But everything has posited that we know the graph somehow. This chapter finally deals with where the graph comes from.