Hobbservations Climate Change, Development, Data

Time for Cows: Learning from Games

I designed a game about cows, droughts, and insurance to play with pastoralists in Kenya. The current version is far from user-friendly, but might be interesting to people engaged in this sort of research. We’ll be using the data to put together a model of their decision-making under risk that we hope will be accurate enough to help us design insurance products that better meet their needs.

Simulating Daisies with D3

Daisyworld is a simulation of a planet that is slowly getting hotter, and I thought it seemed like a fun way to experiment with D3.js. It is inhabited by two species of daisies: dark daisies and light daisies. Dark daisies absorb solar radiation, meaning they can grow in cool temperatures and help the planet get warmer. Light daises reflect solar radiation and help the planet cool off. Without any coordination, dark and light daisies can keep their planet at a habitable temperature, even with variable solar radiation.

Earth Engine Econometrics

Google Earth Engine is not built to do econometrics, but it makes doing so with geographic data from a low-powered (and small hard-drived) computer practical. This post details one method to leverage the Earth Engine platform for panel regression, which makes it possible to obtain a plausibly causal estimate of the effects of weather fluctuations.

Scraping Data from Many Big PDFs

The California Carbon Dashboard’s data comes from PDFs downloaded from Intercontinental Exchange (ICE). Unfortunately, these PDFs are daily and are more than 2000 pages long. In order to get the data out of them in a reasonable amount of time, we needed to figure out a way of quickly identifying the right pages and pulling tables from them. I figured I’d document this here, as I think it might be useful to someone else’s research.

Using Climate and Weather Data for Economic Analysis

I prepared a slide deck for my Environmental Economics class based on a very useful paper by Auffhammer, Hsiang, Schlenker, and Sobel. I’m posting the slides here so they can be used as a quick reference for myself and others when it comes time to use these data.