Share this page!

Enter URL or ID to Unroll

You can paste full URL like: https://x.com/threadreaderapp/status/1644127596119195649
or just the ID like: 1644127596119195649

How to get URL link on X (Twitter) App

  1. On the Twitter thread, click on or icon on the bottom
  2. Click again on or Share Via icon
  3. Click on Copy Link to Tweet
  4. Paste it above and click "Unroll Thread"!
  5. More info at Twitter Help
Carolina Caetano Profile picture
Carolina Caetano
@carolcaetanoUGA
I’m an econometrician @ECON_Terry (UGA) Endogeneity, discontinuities, bunching, health and early childhood.
Sep 22, 2020 • 11 tweets • 5 min read
If your treatment variable has bunching, this is good news! It may be possible to use this bunching to build a correction for endogeneity. We show how this may be done in our new paper. Follow the thread. #EconTwitter #econometrics 1/ We focus here on bunching at one of the ends of the support, which turns out to be very common. For example, TV watching, enrichment activities, maternal labor supply, smoking amounts, all have bunching at zero, and the list of variables where this happens goes on and on. 2/ Image
Jul 7, 2020 • 8 tweets • 2 min read
I’m very interested to hear any thoughts on this paper. There was a version out in 2014, but seems way more complete now. Should choice of polynomial degree in RDD be endogenous? 1/8 …9-a-62cb3a1a-s-sites.googlegroups.com/site/peizhuan/… Paper: Local Polynomial Ord... In principle, this certainly makes sense. Near the threshold all processes we study are fairly linear. However, as we move out convexities and concavities appear and bias increases. If we did a quadratic instead of linear fit, we could go father away from the threshold. 2/8
Jul 3, 2020 • 4 tweets • 1 min read
Really important work. I worked with the PSID and know that the potential uses go way beyond what is currently being done with those datasets. It’s very hard data to collect well, but extremely useful to understand household decisions. Here is why time use has potential beyond its current uses: bunching. For almost every time use activity there are plenty of people who do zero of it. This bunching can be used to test endogeneity in models and even to correct for endogeneity.
Jul 2, 2020 • 6 tweets • 2 min read
Our paper on the RDD with multiple endogenous variables and a single running variable and threshold is ready. I’m proud of this work with Gregorio Caetano and Juan Carlos Escanciano. arxiv.org/abs/2007.00185 Image We had some versions of this out before, but it took us a while do understand this problem fully. At the essence, it is an impossible problem. We show which assumptions must be done to gain each piece of information. 2/