Zoom Logo

Causal Inference Advanced Workshop Day 3 - Shared screen with speaker view
Praveen Bhagawan
24:36
Good Morning!
Scott Cunningham
24:45
Good morning everyone!
Amit Mehra
30:05
voice not clear over zoom
Scott Cunningham
30:21
Hold on let me tell him
Charles Stoecker
30:43
It's great if he stays behind the podium.
Charles Stoecker
30:49
It's muddy when he's out in front.
Matt Kohnhorst
31:40
The mic/speaker for zoom is on the podium, so that makes sense. Sorry, we'll let him know.
Grace Eau
31:59
Yeah
Kira Vila
32:01
But it also looks like he has a remote mic
José Lima
32:17
I believe the mike is not properly put on
Grace Eau
32:21
Sound is echoing
Scott Cunningham
32:51
Is that working?
José Lima
32:51
thanks
Scott Cunningham
32:53
Okay
Amit Mehra
32:55
Thanks
Praveen Bhagawan
32:59
Thank you
Grace Eau
32:59
Thanks, much better
Leandro Sanz
33:00
Thanks
Hyunwoo Jung
33:04
Thanks 🙂
José Lima
01:21:16
actually you can obtain a distribution of out-of-sample R2 by repeating this exercise many times with different training and testing samples. can you use this information to test for the best f?
Monika Avila
01:21:46
In case of clustering with cross-sectional correlation within clusters, cross-validation can only be done by splitting the data at cluster level ?
Muhammed Altug Turgay
01:30:51
Does anyone have the dropbox link of presentation files?
Scott Cunningham
01:32:02
https://www.law.northwestern.edu/research-faculty/events/conferences/causalinference/readings/
Scott Cunningham
01:32:05
Click on 2022 readings
Scott Cunningham
01:32:10
Hansen is listed under Day 8
Monika Avila
01:41:39
does the pooled estimator achieves the semiparametric efficiency bound?
Wei Li
02:20:37
I wonder if there is a reference for the beta estimator on page 26
José Lima
02:44:04
coefficients close to zero in ridge should vanish in lasso?
José Lima
02:49:30
variables should normalize before applying penalty models? variables with different scales should have different coefficient scales as well.
José Lima
02:58:29
can we move the center of penalization of some variables away from zero?
Syed Waleed Mehmood Wasti
04:43:36
Sound is out
Lev Klarnet
04:44:18
no sound
Amit Mehra
04:48:01
no sound
Amit Mehra
04:48:31
it seems to be fading in and out
Xiaoying Yu
04:50:29
it seems to be fading in and out+1
Matias Martinez
04:50:37
+1
Matt Kohnhorst
04:54:17
Starting back up now
xuhong Su
05:03:34
Where are his slides?
xuhong Su
05:03:48
He is not opening his PPT slides
Hongyu Fu
05:08:02
Does the order of variables in decision nodes matter on fitting?
Monika Avila
05:24:15
For bootstrapping, we need iid data ?
Monika Avila
05:52:09
What is depth? I did not hear
Bernard Black
05:57:50
The number of levels in the tree
Monika Avila
05:58:05
Thank you
Rahul Suhag
06:06:22
any good resource that talks about neural nets for image/audio data?
Scott Cunningham
06:51:33
that's so crazy
José Lima
06:52:20
a nice haircut can make the difference
Kira Vila
07:32:59
okay, that's kinda/really cool
Kira Vila
07:34:45
could sample splitting then be a robustness check for overfitting?
Scott Cunningham
07:58:10
Everyone< I have to run to catch my plane. But I just wanted to say thank you for attending and good luck with everything.
Monika Avila
07:58:33
Thank you for the organization!
Monika Avila
07:58:51
Have a nice flight!
Kira Vila
08:12:22
Thank you!! This was a great event!
Amit Mehra
08:13:32
Thanks - it was a great learning experience!
Hongyu Fu
08:14:14
Thank you for organizing this amazing workshop.