11 Oct 2017
Data Science Philippines: Causal Inference and Big Data
This talk provides a high-level, fairly non-technical introduction to causal discovery using big data; i.e., how to carefully draw causal conclusions from big data analyses. Two general, complementary approaches for causal discovery will briefly be illustrated in the context of big data analysis: 1.) mechanism-focused and structural approaches using causal graphs, and 2.) the effect-focused statistical framework of potential outcomes (emphasis on the latter).
15 Mar 2017
Three Statistically Significant Principles
This is a short presentation I gave at the Quantified Self Bay Area Meetup event titled "Show & Tell #41" on March 15, 2017. Summary:
Big data does not imply big accuracy. (S)
Significance does not imply importance. (S)
Correlation does not imply causation. (P)
Causation can imply correlation. (P)
A Statistical New World
Some excellent grad school friends of mine created this fun musical take (based on Disney's "A Whole New World") on what it's like to be a statistician---and asked me to perform and handle music production! From the 2011 American Statistical Association "Promoting the Practice and Profession of Statistics" Video Competition.