12:15:26	 From  Lita-NBER Staff : Welcome to the Summer Institute HE Workshop
12:16:42	 From  NBER Conference  to  Lita-NBER Staff(Privately) : keep yourself muted
13:47:22	 From  Benjamin Hansen : Thanks Mike!!!  
13:47:54	 From  chrisruhm : Thanks Mike. What a joy it has been working with you all these years.
14:09:08	 From  Pragya Kakani  to  Lita-NBER Staff(Privately) : Hi Lita - sorry I got disconnected and logged back in. I think my co-host status is gone though. Could you make me a co-host again? My co-author Adam is presenting at 3PM.
14:15:16	 From  Kitt Carpenter/Nashville/he, him : Is restricting to whites just because of small samples of nonwhites?
14:15:57	 From  Nick Wilson : (My question/statement too: strange that survey seems to have only sampled whites for this.  Or this is a researcher restriction?
14:16:31	 From  Jason Fletcher : Restricted by researcher.  Sample is nationally representative
14:18:11	 From  James Atticus Bolyard : The score was constructed based on only individuals with European ancestry. The reasons for stratifying analysis by race is that there are concerns that a score could pick up SNP's unique to priveleged groups as SNP's predicting education. These genetic effects would not actually reflect genetic propensity for education but instead unequal access to educational resources based on race.
14:20:25	 From  Kitt Carpenter/Nashville/he, him : In ten minutes we will open the Q&A. If you would like to be in the queue, please send me a chat to indicate so. (I do not see the raise hand feature.)
14:22:35	 From  Ce Shang : How is the PGS developed? Could there be same genetic information that can explain both educational attainment and health so that they could be the component of both the PGS for educational attainment and the pigs for health
14:23:40	 From  Kitt Carpenter/Nashville/he, him : If you are not a co-host you should be able to see the 'raise hand' feature, which will also tell me to put you in the queue.  (thanks david slusky)
14:26:45	 From  James Atticus Bolyard : The PGS is derived empirically by regressing all available SNP's on a  phenotype of interest (years of education). Scores are then created by aggregating the coefficients of only statistically-significant SNP's for each individual's unique genetic information (and controlling for SNP's that are highly correlated to avoid double-counting). 
14:28:52	 From  Nick Wilson : Why is there no score for groups other than whites?  I still do not understand.  You explained why stratified by race, but you did not explain why only whites.
14:38:57	 From  Gabriella Conti : I was wondering whether the PGS are constructed also for the siblings and if you use this?
14:41:53	 From  Douglas Staiger : Have the PGS been used to explore heterogeneous treatment effects in randomized trials? e.g., one of the conclusions is that improving SES may have particularly important health effects for high PGS (due to interaction) -- so that could be tested in a trial like MTO or income support that tries to improve SES.
14:42:12	 From  John Cawley : To build on Gabriella's question: an appealing thing about using just siblings is that Mendelian randomization is conditional on the same two parents.
14:42:59	 From  Matthew Lindquist : Does a child's PGS predict their parents' SES?
14:43:36	 From  David Slusky : Can you do this the other way?  Meaning, calculate a PGS for health using the same method and then run your analysis for education?
14:45:38	 From  Rosalie Pacula : The PGS weights are constructed from an outside sample, or inside Add Health data sample?  If it is outside sample, have you considered the association between PGS and highest educational attainment of the actual students you are looking at Add Health , particularly by SES?? I don't see it in the draft paper on-line.
14:49:49	 From  Gabriella Conti : Have you tried to reproduce these densities with the non-white sample? (Back to the qs on race)
15:25:26	 From  Kitt Carpenter/Nashville/he, him : Reminder to please raise your participant hand or chat me to get in the queue, which begins in about 5 mins.
16:08:19	 From  Nicolas Ziebarth : sorry maybe I missed it but could you give us an idea about the APL threshold in India relative to the average income? There is huge heterogeneity in income in rural vs. urban regions right? Is this one APL like 100% FPL in the US? Thanks.
16:11:14	 From  Cynthia Kinnan : BPL is roughly the bottom quartile. In Karnataka the threshold is Rs. 900/person/mo in rural and 1100/person/mo in urban.
16:11:30	 From  Cynthia Kinnan : So APL is above that threshold.
16:11:48	 From  Nicolas Ziebarth : THANKS, that is very helpful!
16:12:49	 From  Nicolas Ziebarth : AH now it is explained. Sorry!
16:25:12	 From  Kitt Carpenter/Nashville/he, him : Reminder to please raise your participant hand or chat me to get in the queue, which begins in about 5 mins.
16:31:27	 From  David Slusky : Does anyone run out of funds?  Do you still utilization drop when benefits are done?
16:31:41	 From  David Slusky : (*see* not *still*)
16:35:14	 From  Lisa Schulkind : I was surprised that only 25% of untreated ailments were untreated due to affordability. Did I remember/understand that statistic right? and if so, do you think that might have something to do with why people don’t change their behavior much once the affordability barrier is removed?
16:35:37	 From  Cynthia Kinnan : Unfortunately we haven't been able to get claims data, so we can't precisely see if/when people run out of funds. That would be interesting to look at, though.
16:49:14	 From  Nick Wilson : Uh, then first stage is strong.  Puzzling that 2-stage is underpowered.  I guess I misunderstood.
16:53:40	 From  Benjamin Hansen : My question was that the most consistent finding of the Oregon Health Insurance experiment was that it improved financial well being.  Which is the first order goal of an insurance to reduce financial risk.  Do you have any findings on financial well being and what are the challenges to measuring that in developing countries? 
16:58:15	 From  Cynthia Kinnan : In terms of financial well-being, yes, that's what we're looking at currently. We hope to have a draft on those outcomes ready to post/circulate in the not-too-distant future!
16:59:30	 From  Cynthia Kinnan : In terms of challenges, yes, one challenge is measurement; another is that there are many margins that people can change along (asset holdings, spending, loans, gifts among households, etc.).
18:09:31	 From  Kitt Carpenter/Nashville/he, him : We love you Mike and Ilene
18:10:35	 From  Inas Kelly : Yes - we love you, Mike and Ilene!
18:11:11	 From  Anca Grecu : Thank you Mike!
18:11:11	 From  Benjamin Hansen : Thank you for everything Mike and Ilene!!! Can't wait to be able to visit you in NYC sometime (hopefully soon)