Statistical inference | Causal inference | Design of experiments | Survey methodology | Asymptotic analysis
In experiments, a spillover is an indirect effect on a subject not directly treated by the experiment. These effects are useful for policy analysis but complicate the statistical analysis of experiments. Analysis of spillover effects involves relaxing the non-interference assumption, or SUTVA (Stable Unit Treatment Value Assumption). This assumption requires that subject i's revelation of its potential outcomes depends only on that subject i's own treatment status, and is unaffected by another subject j's treatment status. In ordinary settings where the researcher seeks to estimate the average treatment effect, violation of the non-interference assumption means that traditional estimators for the ATE, such as difference-in-means, may be biased. However, there are many real-world instances where a unit's revelation of potential outcomes depend on another unit's treatment assignment, and analyzing these effects may be just as important as analyzing the direct effect of treatment. One solution to this problem is to redefine the causal estimand of interest by redefining a subject's potential outcomes in terms of one's own treatment status and related subjects' treatment status. The researcher can then analyze various estimands of interest separately. One important assumption here is that this process captures all patterns of spillovers, and that there are no unmodeled spillovers remaining (ex. spillovers occur within a two-person household but not beyond). Once the potential outcomes are redefined, the rest of the statistical analysis involves modeling the probabilities of being exposed to treatment given some schedule of treatment assignment, and using inverse probability weighting (IPW) to produce unbiased (or asymptotically unbiased) estimates of the estimand of interest. (Wikipedia).
Test done with Blender 2.5. http://www.kostackstudio.de
From playlist Random Blender Tests
Testing and Online Experimentation
Join Data Science Dojo and Statsig for a conversation on experimentation and testing. Learn how leading companies like Facebook use experimentation to build better products and accelerate their growth with 10x as much testing. Web experimentation can range from simple projects like design
From playlist A/B Testing & Beyond
Blender test: Raindrops on a water surface
This is much more complex than it seems. The mesh isn't very detailed, but the fully animated normal and displacement textures are. They were created by a python script beforehand. And guess what, they are seamless and loopable! :) http://www.KaiKostack.com
From playlist Random Blender Tests
AWESOME Physics demonstrations. Interference of water waves experiment.
A ripple tank is placed above a mirror and a projection screen. Two synchronous point sources, whose frequency can be varied, tap the surface of the water and produce circular waves. The interference pattern of the waves including the lines of nodes can be observed on the screen.
From playlist WAVES
Here is a demonstration of the doppler effect.
From playlist All Demonstrations
MIT 14.771 Development Economics, Fall 2021 Instructor: Ben Olken View the complete course: https://ocw.mit.edu/courses/14-771-development-economics-fall-2021 YouTube Playlist: https://www.youtube.com/playlist?list=PLUl4u3cNGP61kvh3caDts2R6LmkYbmzaG Continues discussion of labor, with
From playlist MIT 14.771 Development Economics, Fall 2021
Ses 2 | MIT Abdul Latif Jameel Poverty Action Lab Executive Training
Session 2: Why randomize? Speaker: Dan Levy See the complete course at: http://ocw.mit.edu/jpal License: Creative Commons BY-NC-SA More information at http://ocw.mit.edu/terms More courses at http://ocw.mit.edu
From playlist MIT Abdul Latif Jameel Poverty Action Lab Executive Training
Policy Priority Inference: Simulations for Government Strategy | AISC
For slides and more information on the paper, visit https://ai.science/e/synthetic-data-policy-priority-inference-simulations-for-government-strategy--9wTcGo34oeeckPY6ncLz Speaker: Omar Guerrero, PhD ; Host: Chenda Bunkasem Motivation: This research shows how synthetic data can be used
From playlist Synthetic Data
Replication, Re-Analysis, and Worm Wars
A number of you have asked for an episode on worm wars. Others of you will have literally no idea what worm wars is. This episode is for both groups. Worm wars are the topic of this week's Healthcare Triage. Those of you who want to read more can go here: http://theincidentaleconomist.c
From playlist Healthcare Triage
5.2 - Pathogen evolution: Virulence 2
"Evolutionary Medicine" Sinauer Associates (2015) is the textbook that supports these lectures. Instructors can request examination copies and sign up to download figures here: http://www.sinauer.com/catalog/medical/evolutionary-medicine.html
From playlist Evolution and Medicine (2015) with Stephen Stearns
take a peak at exploratorium exhibits!
Check out cool on-line exhibits at http://www.exploratorium.edu/explore
From playlist Hands-on Exploratorium
Trailer - Early Attempt: WTC-7 Collapse Simulation (obsolete, check the improved version of 2017)
Latest version: https://youtu.be/VAkTbyENZ5s Short clip of my full case study can be found here: https://www.youtube.com/watch?v=MSlIFXw3EGg Made in Blender.
From playlist Random Blender Tests
Miles to go before I sleep: gamification, health wearables and activity
Health wearables in combination with gamification enable interventions that have the potential to increase physical activity—a key determinant of health. However, the extant literature does not provide conclusive evidence on the benefits of gamification and there are persistent concerns th
From playlist Social and Ethical AI
Ses 7 | MIT Abdul Latif Jameel Poverty Action Lab Executive Training
Session 7: Managing threats to evaluation and data analysis Speaker: Michael Kramer See the complete course at: http://ocw.mit.edu/jpal License: Creative Commons BY-NC-SA More information at http://ocw.mit.edu/terms More courses at http://ocw.mit.edu
From playlist MIT Abdul Latif Jameel Poverty Action Lab Executive Training
Solar panel performance shoot-out - Part 2
This is a performance test between two 55 watt solar panels, one is a mono-crystalline and the other is an Amorphous / thin film panel.
From playlist Solar Panel Reviews, Testing and Experiments
Ses 3 | MIT Abdul Latif Jameel Poverty Action Lab Executive Training
Session 3: How to Randomize I Speaker: Steve Harvey See the complete course at: http://ocw.mit.edu/jpal License: Creative Commons BY-NC-SA More information at http://ocw.mit.edu/terms More courses at http://ocw.mit.edu
From playlist MIT Abdul Latif Jameel Poverty Action Lab Executive Training
Simon Frost: Towards reproducibility and transparency in mathematical and computational epidemiology
Mathematical models of infectious disease transmission are increasingly used to guide public health and policy decisions. Hence, it is important that every effort is made to ensure that models are ‘correct’, made difficult by the frequent need to simulate a model numerically. The best we c
From playlist Combinatorics
New study shows that liquid droplets form in unexpected ways. Read more: http://bit.ly/2d4Psgn JOIN AAAS: http://scim.ag/2bxrxnH
From playlist Materials and technology
Policy Priority Inference for Sustainable Development, Omar Guerrero
Guidelines to support international development of developing countries are many. Today, the best example of these guidelines is the Sustainable Development Goals (SDGs) – a set of 17 general goals monitored through 232 development indicators. Achieving these goals is, however, a complex t
From playlist Driving data futures
20 AMAZING SCIENCE EXPERIMENTS compilation!!!
This video is a compilation of Best 20 science experiments with liquid, fire, laser, mechanics, electricity and waves.
From playlist WAVES