Profile avatar
pbousquet.bsky.social
Econ PhD @ UVA | mostly sharing macro research from twitter until people are more active here | pbousquet.com
21 posts 144 followers 869 following
Regular Contributor
Conversation Starter
Posts 21 Comments 18

Bayesian logic in entrepreneurial experimentation continues to apply under Knightian uncertainty, from @joshgans.bsky.social https://www.nber.org/papers/w33507

submitted 2 days ago • 0 comments

Deriving asymptotically unbiased estimates of the multi-step forecasting risk and the impulse response estimation risk to determine hyperparameters in settings where the Bayesian vector autoregressions is (potentially) misspecified, from González-Casasús and Sch... https://www.nber.org/papers/w33474

submitted 11 days ago • 0 comments

I couldn't agree more! The excessive focus on linearized solutions to HA models is a bit like the old joke about the drunk who is looking for a key under a lamppost because that's where the light is. More discussion of why we need non-linear models and ways forward here benjaminmoll.com/challenge/

submitted 14 days ago • 1 comment

THREAD 🧵: A Concerning Trend in the Heterogenous-Agent (HA) Macro Literature 1/n A growing class of computational algorithms makes heavy use of linearization methods to solve HA models. Two key benefits are emphasized: (i) (extreme) speed & (ii) allows many state (aggregate) variables. ++ #econ

submitted 15 days ago • 6 comments

Some machine learning methods — like "old-school" Ridge — outperform sparsity-oriented approaches in leveraging weak signals for economic prediction, from Zhouyu Shen and Dacheng Xiu https://www.nber.org/papers/w33421

submitted 25 days ago • 0 comments

Another good one from Plagborg-Møller and co! #EconSky

submitted 25 days ago • 1 comment

New post on Substack about tackling p-hacking in science. P-hacking—practices that inflate the chance of significant results—is pervasive and threatens scientific progress. How can we model it and then correct for it? Read here: pmichaillat.substack.com/p/how-can-we...

submitted 43 days ago • 0 comments

Modern supply chains don't look like trade theory 101! They involve constant border crossings, each now hit by tariffs. Tariffs raise prices, but the more important thing they do is disrupt supply relationships. 1/

submitted 29 days ago • 37 comments

One of the wildest facets of the last couple years is that not only are we seeing a stealth deleveraging of consumers' housing balance sheet (subdued origination -> amortization rolls along and fixed payments shrink vs rising incomes) but also an outright deleveraging for non-housing debt too.

submitted 53 days ago • 7 comments

Thanks @davideromelli.bsky.social for making this great data set available: "The dataset features 35,487 unique speeches from 131 central banks, for the period going from the beginning of January 1986 (date of the first online speech by a central bank) to the end of December 2023."

submitted 41 days ago • 0 comments

My traditional end-of-year review: some papers I read and liked in 2024. donskerclass.github.io/post/papers-...

submitted 62 days ago • 2 comments

We concluded our first round interviews this week for a position in econometrics. Wow, I'm glad I was on the market a long time ago. So many brilliant young minds. Some relatively obvious themes in topics: #econsky

submitted 80 days ago • 1 comment

Want to solve heterogeneous-agent models with multiple endogenous state variables, say a multi-asset HANK model? @soroushs.bsky.social and @patricksecon.bsky.social developed a much-improved finite-difference scheme for such problems: github.com/SSabet/Neste... Also here benjaminmoll.com/codes/

submitted 74 days ago • 2 comments

I didn't know this talk was being recorded, but the videos for all the ACM-EC workshops are online! Check the rest of the session out too for some very helpful computational macro tips. youtu.be/XYGR3lnbK34?...

submitted 83 days ago • 1 comment

Thanks to @donskerclass.bsky.social I found that recordings of our ACM EC-24 talks are online. In case you are interested, my overview-ish talk on deep learning based solution methods (focus on aggregate risk and OLG) is available here: youtu.be/CPQpXc5PCJU?... #econsky

submitted 77 days ago • 0 comments

Hard to overstate how revelatory this paper was for me — one takeaway I had from it was "small fixed costs can really deter people from optimizing." Which made me think about a few other recent papers where fixed optimization costs are a key ingredient:

submitted 83 days ago • 3 comments

Now that @benmoll.bsky.social has gotten everybody's attention, let me tell you about the JMP of my brilliant student Michael Cai, which deals directly with Ben's challenge: How can we model expectations in HA models in a way that is both tractable and consistent with the evidence? #EconSky

submitted 88 days ago • 1 comment

New paper: benjaminmoll.com/challenge/

submitted 93 days ago • 3 comments

My friend and colleague Joe Anderson is on the market with an incredible #EconJMP on equilibria when government debt is issued in a way that conflicts with the central banks' objectives. One of those papers where after reading you see all relevant news through its lens! #EconSky (🧵 below)

submitted 97 days ago • 1 comment

New working paper from Kolesár and Plagborg-Møller is a must read for anyone using VARs or LPs (so pretty much everyone in macro these days..)! #EconSky

submitted 104 days ago • 1 comment

#EconSky Possibly biased because I am already firmly in this camp, but good new paper making the case for economists to make heavy investment into deep learning. Hope to share my experiences using PyTorch to solve highly non-linear models soon www.galonuno.com/uploads/1/3/...

submitted 115 days ago • 1 comment