Similar User
@u02lilmx_C
@aki_asa_
@seihkuya
@EagletalonQ
@HighChewww
@suwako009
@chunochunoChuno
ブログ更新しました。 / 『ベイズデータ解析』はベイズ統計学を用いる全ての実務家が座右に置くべき第一級の鈍器 - 渋谷駅前で働くデータサイエンティストのブログ tjo.hatenablog.com/entry/2024/08/…
News on sdid_event: - SDiD, DiD and SC can all be computed via the method() option; - added placebo() option to estimate pre-trends; - the package is now on SSC, so check it out by running ssc install sdid_event, replace Many thanks to @DanielPailanir and Damian Clarke!
年齢制限ギリギリで賞をいただきました www2.econ.tohoku.ac.jp/~DSSR/hosoya20… そろそろ色々と若手枠から外れるの悲しい😧
One hard thing about seeing the Callaway and Sant’Anna DiD estimator becoming so popular is to see people using it in a context that is not warranted. Please read the paper and understand its building blocks! But I guess these are rich people's problems!
RDD、というかその使い方に警鐘を鳴らす論文 Can Close Election Regression Discontinuity Designs Identify Effects of Winning Politician Characteristics? - Marshall - American Journal of Political Science - Wiley Online Library onlinelibrary.wiley.com/doi/10.1111/aj…
Sneak peak of some slides from a guest lecture in a colleague's grad class on economic modeling. I had a lot of fun with these -- might try to write it up at some point
Recently accepted in REStud, ``Partially Linear Models under Data Combination,'' from D'Haultfoeuille, Gaillac and Maurel: restud.com/partially-line…
Hyvarinenスコアを使って非正規化カーネル密度推定するとハイパラがうまくチューニングできて柔軟な推定ができますよという論文がJBESにアクセプトされました!! 京大M2の今井さんが理論パートをゴリゴリ進めてくれました!! arxiv.org/abs/2212.00984
Classical p-values and the Bayesian posterior probability that the hypothesis is approximately true dlvr.it/T2rtQz
Biometrika 採択論文 (Open Access article) 「Explicit solutions for the asymptotically-optimal bandwidth in cross-validation」 Our bandwidth formulae are simple and noniterative thus leading to very fast computations … doi.org/10.1093/biomet…
本を書くことによる金銭面でのメリットはそこまでないけど、若手研究者のみなさんに学生のときに本を読んで勉強しましたと言われるのはうれしい。
On the failure of the bootstrap for Chatterjee's rank correlation. Zhexiao Lin, Fang Han. Biometrika. doi.org/10.1093/biomet…
New paper! TWFE for diff-in-diff is biased under staggered adoptions. One fix is to use more parameters--estimate a separate treatment effect for each cohort at each time. But treating each TE like a free parameter wastes our knowledge that nearby TEs in time are likely similar.
"Gravity estimations with Staggered Difference-in-differences" ideas.repec.org/p/ris/drxlwp/2… Most credit for our new working paper goes to my brilliant co-author Arne J. Nagengast @bundesbank Thanks to all colleagues who gave us great comments and suggestions! @EconTwitter
横からすいません．ガウス過程事前分布は対応するRKHSが無限次元ならその標本路は確率１でそのRKHSに"含まれない"です．なので，事前分布固定のガウス過程回帰で最適レートを出そうと思ったら，対応するRKHSより広い範囲のクラス（より粗い関数）を考える必要があります． jmlr.csail.mit.edu/papers/volume1…
Confidence intervals for the Cox model test error from cross‐validation. Min Woo Sun, Robert Tibshirani. Statistics in Medicine. onlinelibrary.wiley.com/doi/10.1002/si…
United States Trends
- 1. #TSBlackFridayCollection 2.018 posts
- 2. $DCK N/A
- 3. Thanksgiving 186 B posts
- 4. $CUTO 14 B posts
- 5. Kreider 2.206 posts
- 6. #BBMAs 156 B posts
- 7. Axios 20,3 B posts
- 8. #SonicMovie3 58,5 B posts
- 9. #MondayMotivation 22,2 B posts
- 10. #IDontWantToOverreactBUT 1.272 posts
- 11. $FRED 23,1 B posts
- 12. Trouba 1.660 posts
- 13. Porzingis 3.280 posts
- 14. Phylicia Rashad 2.303 posts
- 15. Outlook 11,1 B posts
- 16. Kemp 4.476 posts
- 17. Rumble 30,9 B posts
- 18. Marshall Law 2.393 posts
- 19. Kikuchi 7.380 posts
- 20. Victory Monday 4.523 posts
Something went wrong.
Something went wrong.