iiinova — where your digital self keeps evolving

Rachel Leah Childers @donskerclass.bsky.social · 30/09/2026

What I really want to ask the Keynes bot:

Rachel Leah Childers @donskerclass.bsky.social · 10/09/2026

I used to give a problem set to my students using this guinea pig DAG to teach d-separation, backdoor criterion, etc. It's not nearly so charming when encoded in dagitty and plotted in standard ggdag style - someone clearly needs to add a theme_guinea_pig()

Sewall Wright's guinea pig DAG from the parent post, encoded in R library Dagitty and plotted in R library ggdag with black circles for nodes and letters for labels rather than the charming hand-drawn guinea pig nodes in the original.

071

Rachel Leah Childers @donskerclass.bsky.social · 31/07/2026

This is exactly the kind of thing I was looking for, thanks! I will try out the code and read the paper, which seems to have a comprehensive and up-to-date set of comparisons.

Figure 1 from "Particle-MALA and Particle-mGRAD: Gradient-based MCMC methods for high-dimensional state-space models" by Corenflos and Finke

The graphs compare expected squared jumping distance (a performance measure) with problem size parameters (dimension D of state and time horizon T) for 13 different variant state space samplers, on a toy linear-Gaussian state space model.

Graph (a) Empirical scaling with D for fixed time horizon T = 25. Graph (b) Empirical scaling with T for fixed state dimension D = 10.

110

Rachel Leah Childers @donskerclass.bsky.social · 25/07/2026

OMG, someone found Yu Deng's Xiaohongshu post about the win. I love... PDE 😭

Manga drawing of a blonde girl in a school uniform with the word "PDE" over her head, looking back at a dark-haired girl in the same uniform, labeled "2022 Fields Medal" running away

Manga drawing of blonde girl with "PDE" over her head embracing and about to kiss dark haired girl with "2026 Fields Medal" over her head.

030

Rachel Leah Childers @donskerclass.bsky.social · 23/07/2026

Inspirational: one of the new Fields Medalists is a himedanshi.

For comfort, Deng turned to manga. “I remember him at Courant. He was always holding a manga,” said Jalal Shatah(opens a new tab), a professor there who formerly chaired the math department. Deng particularly liked stories about deep friendship and romance and found himself drawn to a genre known as yuri, which focuses on those kinds of relationships between women — stories that “feel nice and warm and beautiful,” as he put it. They helped him be kinder to himself, “more comfortable with life.”

26122

Rachel Leah Childers @donskerclass.bsky.social · 25/06/2026

New answer to "Which moments to match?" just dropped, by Andrews and Sanders economics.mit.edu/sites/defaul... Pick parameter value \bar{theta} to fit your ideological goal, and then match moments in the following set, and you are guaranteed to get the result you want. 😬

$The set of moment functions f(X) you can choose for the method of moments applied to a model with likelihood P_{\theta}, which will ensure that the parameter value which matches the moments is an arbitrary value \bar{\theta} chosen by the researcher is given by the set of all square integrable functions which are orthogonal to the linear span of the likelihood ratio of the true distribution P with respect to the model-implied distribution P_{\bar{\theta}} (after subtracting 1 from the likelihood ratio for centering). This set is nontrivial precisely when the model is mis-specified, meaning that the likelihood ratio of the data distribution with respect to the model implied distribution at \bar{\theta} is not equal to 1, with positive probability.$

1156

Rachel Leah Childers @donskerclass.bsky.social · 15/06/2026

A helpful clarifying note from Greg Kaplan on uniqueness in the New Keynesian model under different fiscal and monetary rules, and comparison with the non-Ricardian case. Continuous time gives closed forms that simplify the analysis a lot. static1.squarespace.com/static/5d694...

Phase diagrams of the Representative Agent New Keynesian model in continuous time in the passive monetary active fiscal regime, linearized on the left and global (under Rotemberg pricing) on the right. In both there is a saddle point normalized values of output Y=1, inflation pi=0. There is a 1d stable manifold in the Y,pi space which converges to this saddle. For a given initial debt, any point (y,pi) on this submanifold is stable. However, a government debt equation, which evolves autonomously from the (Y,pi) subsystem with Ricardian consumers, requires stable values of these variables and so selects the saddle point as unique initial condition in order to not blow up over time, a condition known as the fiscal theory of inflation (which in discrete time selects the price level, hence the "Fiscal Theory of the Price Level"). Kaplan notes that this selection is implied by transversality for the consumers holding the debt, so is a unique equilibrium at least under a boundedness restriction for (Y_t,pi_t) and not an external equilibrium refinement.

In the paper he discusses why economists other than John Cochrane have been reluctant to consider this setting as plausible, and discusses the non-Ricardian case where B enters the Euler equation and picks out a similar solution under purely local criteria, though global uniqueness is still only a conjecture for that case.

131

Rachel Leah Childers @donskerclass.bsky.social · 18/04/2026

Springtime Saturday skateboarding. 🛹

Me on a skateboard, with hair in my face despite the hair tie.

Me riding on a skateboard near some train tracks.

0130

Rachel Leah Childers @donskerclass.bsky.social · 02/04/2026

I found a copy in a used book store several years ago and picked it up solely to make a Linear Algebra joke about the title. Anyway, I'm a woman now.

Cover of "The Identity Matrix" by Jack L. Chalker
The cover art is a drawing of a woman below a UFO-style spacecraft.

Back Cover of "The Identity Matrix"

Text: The Invaders are among us

"While backpacking in Alaska, Victor Ganser, a 35-year-old Ph.D. in political science, finds himself trapped in the body of a 13-year-old Tlingit Indian girl. A little later he is forced from there into the spectacularly beautiful body of an 18-year-old blonde. He finds that the switches are part of a skirmish for Earth by two alien races... And that's just the beginning. Like many of Jack Chalker's earlier books, this one is almost totally unpredictable and will keep you guessing to the last page."

120

Rachel Leah Childers @donskerclass.bsky.social · 26/03/2026

Paul Rosenbaum's Causal Inference book is strongly recommended even if you think you don't need another intro causal inference book. It's short, it's precise, and it's thoughtful about sensitivity analysis and using all the evidence we have. www.goodreads.com/review/show/...

Cover of "Causal Inference" by Paul R. Rosenbaum.

1253

Rachel Leah Childers @donskerclass.bsky.social · 17/03/2026

I am very much not an expert and haven't located the original, but the passage is included and extended in an additional secondary source with commentary and references: rboustan.scholar.princeton.edu/sites/g/file...

Selection from Ra'anan Abusch, "EUNUCHS AND GENDER TRANSFORMATION: Philo's exegesis of the Joseph narrative" in EUNUCHS IN ANTIQUITY AND BEYOND. Shaun Tougher, ed. 2002

p105

Eunuchs and castration in Philo

Philo include castrates among a lengthy list of different types of male
gender-bending and consistently censures them as transgressors of divinely ordained gender categories. In one colourful and strident passage, Philo writes:

In former days the very mention of (pederasty) was a great disgrace, but now it is a matter of boasting nor only to the active partner but to the passive

Abusch p106, continued, quoting Philo Judaeus

partners, who habituate themselves to endure the disease of effemination. They let both body and soul run to waste, and leave no
ember of their male nature to smoulder. Ivfark how conspicuously they braid and adorn their hair, and how they scrub and paim their £'lees with cosmetics and pigments and the like, In tact, the transformation of the male nature to the female is practiced by them as an aft and does not raise a blush". Certainly you may see these hybrids of man and woman continually strutting about through the thick of the market, heading the processions at the feasts, appointed to serve as unholy ministers of holy things, leading the mysteries and initiations and celebrating the rites of Demeter. Those of them who, by way of heightening still further their youthful beauty, have desired to be completely changed into women and gone on to mutilate their genital organs, are clad in purple like signal benefactors of their native lands ... , each of them a curse and a pollution of his country.

281

Rachel Leah Childers @donskerclass.bsky.social · 15/03/2026

I emailed him once a few years ago because he'd proposed (see vid) Bayesian ways of thinking about some classic microeconometric results that bring out the economic structure. I didn't get a response, but working through it changed how I do econometrics. www.chamberlainseminar.org/past-seminar...

A scatter plot of wages vs education from the Angrist and Krueger data with points highlighted in a variety of colors, reflecting posterior draws of cluster assignments in a Dirichlet process mixture of regressions. Sims suggested this approach in his lecture on "Sharp Econometrics", which brought his ideas about really thinking about what goes into the structure of data to microeconometric problems.
See slides and video: https://www.chamberlainseminar.org/past-seminars/spring-2023#h.e0uu7eildjmq

031

Rachel Leah Childers @donskerclass.bsky.social · 13/03/2026

Seconding Cheryl Misak on Frank Ramsey; deeply researched explanation of interweaving themes in his influential work on philosophy, econ, and math and a fascinating character to boot. On the opposite end, Carr on Candy Darling for the 1960s art, celebrity, and NYC social scene.

Cover of "Candy Darling: Dreamer, Icon, Superstar" by Cynthia Carr

020

Rachel Leah Childers @donskerclass.bsky.social · 12/03/2026

My JMP was full of wavelets (mostly Coiflets), and it scared the crap out of several seminar audiences. A whole dictionary of them are available at "Where is the Starlet?" www.laurent-duval.eu/siva-wits-wh...

Coiflet of order 2, from https://en.wikipedia.org/wiki/Coiflet
Coiflets are also due to Ingrid Daubechies, but she gave them to Ron Coifman and let him name them after himself I guess because she also had her own wavelet family.

Wavelet and Chirplet representations of a periodic complex-valued function. Taken from WITS http://www.laurent-duval.eu/siva-wits-where-is-the-starlet.html

260

Rachel Leah Childers @donskerclass.bsky.social · 11/03/2026

Obviously most of the time series arguments about cell phones have hopeless age-period-cohort problems. How many have been cross-checked with staggered geographical introduction? Do we have any credibly identified effects of cell phones on anything other than sardine price volatility in Kerala?

Figure IV, Prices and Mobile Phone Service in Kerala, from
R. Jensen
The digital provide: information (technology), market performance, and welfare in the south Indian fisheries sector
Q. J. Econ., 122 (3) (2007), pp. 879-924 https://doi.org/10.1162/qjec.122.3.879

The figure shows price time series for 3 regions in Kerala in 2001 Rs./kg average 7:30-8:30am beach price for average sardines over 250 survey weeks. Each region is marked with a vertical line showing the date at which phones are added to the market, with introduction around week 20 for Region I, week 100 for Region II, and week 200 for Region III. In all regions, the volatility decreases markedly, rapidly, and persistently after the date of introduction, with price ranges previously varying at high frequency in the 0-14 Rs/kg range subsequently varying by only a few Rupees from week to week.

120

Rachel Leah Childers @donskerclass.bsky.social · 10/03/2026

I read and mostly liked @maxkasy.bsky.social's book on the economics of AI, "The Means of Prediction". I did think it could use a bit more Herbert Simon. Long review: 👇 www.goodreads.com/review/show/...

Cover of "The Means of Prediction: How AI Really Works (and Who Benefits)" by Maximilian Kasy

0100

Rachel Leah Childers @donskerclass.bsky.social · 13/02/2026

This week's reading: Shon Faye's "The Transgender Issue" If I could recommend cisgender people to read one single book about us, it would probably be this one. It's not personal, it's not prurient, there's no philosophy or metaphysics; it's just facts and policy. www.goodreads.com/review/show/...

Book cover: "The Transgender Issue: Trans Justice is Justice for All" by Shon Faye

230

Rachel Leah Childers @donskerclass.bsky.social · 03/02/2026

Never have I felt more like my job will soon by taken by AI. Statistical learning theory in Lean: concentration inequalities, Dudley's entropy integral, and local Gaussian complexity bounds. 30000 lines of code, over 1000 lemmas, formalizing Wainwright and Boucheron et al arxiv.org/abs/2602.02285

Fig 4 from Zhang, Lee, Liu "Statistical Learning Theory in Lean 4: Empirical Processes from Scratch"
The dependency graph of the formalizations. Diagram shows proof of Dudley's entropy integral with preceding lemmas and Gaussian Lipschitz concentration likewise, feeding into a Gaussian Complexity inequality and an error bound for critical radius then used to prove sharp minimax error rates for linear regression.

0335

Rachel Leah Childers @donskerclass.bsky.social · 02/02/2026

For Spring semester, I'm bringing back free weekly open office hours for anyone in the world with Econometrics questions. Tuesdays 4-6 PM Central European Time (US EST 10AM-12PM) or by appointment; sign up and drop by! Details and signup at: donskerclass.github.io/OfficeHours....

32810

Rachel Leah Childers @donskerclass.bsky.social · 29/01/2026

Peeking in on the old site, prominent macroecons are much more active there; they love multi-paragraph posts. Also AI-focused econs, which is less of a loss. Stats is still strong here. But I was never really on Bsky for econ. What are you even doing on the mostly-Katie Tightpussy-by-volume site?

120

Rachel Leah Childers @donskerclass.bsky.social · 18/11/2025

Huh; that delay differential equation has a closed form solution, and it's a perfect cosine curve under the right conditions. Not super surprising, but nice that it works out. doi.org/10.2307/1236...

$From Larson (1964), "The Hog Cycle as Harmonic Motion", some key equations, derived from a continuous time hog cycle model with linear demand and in which quantities respond to prices linearly with a 12 month delay: (4) d/dt p_{t+12} = -bkp_t (5) d/dt q_{t+12} = -bkq_t If bk=1, this system yields a solution (7) p_t = cos(2πt/48 + e_p) (8) q_t = cos(2πt/48 + e_q) where e_p and e_q depend on initial conditions. If they differ by π radians, the solutions are consistent and the system is in resonance .$ Fig 1 from Larson (1964) "The Hog Cycle as Harmonic Motion"

Relationships between the hog-corn price ratio, the numbers of sows farrowing, and pork production, as deviations from trend, 1947-1962.

The chart shows real data for each series, overlaid with harmonic curves of differing amplitude and frequency. The visual fit is fairly good, up to high frequency noise.

Fig 1 from Larson (1964) "The Hog Cycle as Harmonic Motion"

Relationships between the hog-corn price ratio, the numbers of sows farrowing, and pork production, as deviations from trend, 1947-1962.

The chart shows real data for each series, overlaid with harmonic curves of differing amplitude and frequency. The visual fit is fairly good, up to high frequency noise.

000

Rachel Leah Childers @donskerclass.bsky.social · 04/11/2025

Ethel Cain puts on a world-class live show.

Photo of Ethel Cain and band on stage, dramatically backlit, surrounded by branches and hanging vines.

120

Rachel Leah Childers @donskerclass.bsky.social · 28/10/2025

Updated Fall time for free weekly open office hours for anyone in the world with Econometrics questions. Wednesdays 3:30-5:30PM Central European Time (US EDT 10:30/EST 9:30AM) or by appointment; sign up and drop by! Details and signup at: donskerclass.github.io/OfficeHours....

064

Rachel Leah Childers @donskerclass.bsky.social · 01/10/2025

This post by Nan Jiang, nanjiang.cs.illinois.edu/2025/09/29/p... who knows more about RL theory than I ever will, on gaps in standard justifications for Policy Gradient, reassures me as I go through the PG lectures in interactive-learning-algos.github.io that its use is indeed a bit weird.

Office meme from Nan Jiang blog post
Frame 1: Two cards reading "policy gradient with 100 tweaks" and "policy gradient with another different 100 tweaks", respectively. "Corporate needs you to find the differences between this picture and this picture."
Frame 2: Pam Office: "They're the same picture"

010

Rachel Leah Childers @donskerclass.bsky.social · 12/09/2025

I was incredibly glad to catch #jasmine4t on tour last night! They put on an amazing live show, including new tracks from the extended album, out today!

Set list for jasmine.4.t's show in Zurich last night, given to me by the band afterwards

Skin on Skin
Tall Girl
Breaking In Reverse
Find Ur Pppl
I Can't Believe
I Don't Think
Kitchen
You Are The Morning
Tesco
Elephant
Woman
Did U No

020

Rachel Leah Childers @donskerclass.bsky.social · 05/09/2025

Meta-discourse point: under DeGroot learning, signals from multiple sources are reinforced, regardless of whether such information is downstream of a common source (see e.g. stanford.edu/~arungc/CLX....) "Tagging" signals appears to reduce this bias (Möbius et al 2015) but is rarely done.

Figure 7: Information aggregation in naïve model.
From "Treasure Hunt: Social Learning in the Field" by Möbius, Phan, Szeidl
https://www.markusmobius.org/sites/default/files/markusmobius/files/treasure_03_01_2015_0.pdf
Lefthand figure: "No multiple paths" Diagram of directed graph with arrows C->B, B->A, D->A, with root nodes C, D highlighted. Arrows depict transmission of independent signals from B, C, and D to A, which in total are counted once each.
Righthand figure: "Multiple paths" Diagram of directed graph with arrows C->B, B->A, D->A, and now also C->D with nodes C, D highlighted. Arrows depict transmission of independent signals from B, C, and D to A. Because both B and D are downstream of C and upstream of A, the signal from A contributes along both paths. Under naive (DeGroot) learning, signals from correlated sources are not discounted, so A "double counts" the signal from C.
Identities of A, B, C, and D may be filled in by the reader.

120

Rachel Leah Childers @donskerclass.bsky.social · 15/08/2025

I’m pretty sure de basement is a primary input into the cruelty production process.

The Sisters of Dorley
Welcome to Dorley Hall
By Alyson Greaves
Book cover. To spoil the joke, this is a book about a torture basement which I probably shouldn’t be mentioning to cis people.

010

Rachel Leah Childers @donskerclass.bsky.social · 31/07/2025

I haven't looked at the details, but this paper seems to have most of those? joelflynn.com/wp-content/u...

NBER WP 32602, the earlier version of the paper. Formerly titled "The Macroeconomics of Narratives", now "Contagious Business Cycles"

Joel P. Flynn and Karthik Sastry

Abstract: We study the macroeconomic implications of narratives, defined as beliefs about the economy that spread contagiously. In an otherwise standard business-cycle model, narratives generate persistent and belief-driven fluctuations. Sufficiently contagious narratives can "go viral," generating hysteresis in the model's unique equilibrium. Empirically, we use natural-language-processing methods to measure firms' narratives. Consistent with the theory, narratives spread contagiously and firms expand after adopting optimistic narratives, even though these narratives have no predictive power for future firm fundamentals. Quantitatively, narratives explain 32% and 18% of the output reductions over the early 2000s recession and Great Recession, respectively, and 19% of output variance.

131

Rachel Leah Childers @donskerclass.bsky.social · 30/07/2025

Gotta get on the next level (it's full of econometric theory)

Hand-size ringed ruled notebook with Rilakkuma (a Hello Kitt teddy bear) on it, with friends, dressed in a variety of nautical costumes including a seal, an octopus, a ray, and a shark. The contents of the notebook are primarily proofs related to the Generalized Method of Moments, and some derivations about Sobolev spaces.

140

Rachel Leah Childers @donskerclass.bsky.social · 19/07/2025

The title, "Debiased Semiparametric U-statistics" is meaningless to the average practitioner, but indicated it had exactly the theory I needed. It was hard to find because the title is now "Machine Learning Inference on Inequality of Opportunity" about 1 implementable estimator that uses the theory

The latest version of "Debiased Semiparametric U-statistics", with a new title:
Machine Learning Inference on Inequality of Opportunity
Juan Carlos Escanciano, Joël Robert Terschuur
[Submitted on 10 Jun 2022 (v1), last revised 4 Oct 2023 (this version, v3)]
https://arxiv.org/abs/2206.05235

310

Rachel Leah Childers @donskerclass.bsky.social · 19/07/2025

Papers are still coming out trying to fix it to this day. The client's setting was slightly nonstandard, so even those didn't work; doing it right would require combining multiple sources of theory, including results on U-statistics, another ugly area. Luckily, I found a paper w/ general results!

The paper I found on semiparametric U-statistics:
Debiased Semiparametric U-Statistics: Machine Learning Inference on Inequality of Opportunity
Juan Carlos Escanciano, Joël Robert Terschuur
Submitted on 10 Jun 2022 (this version, v1), latest version 4 Oct 2023 (v3)]
https://arxiv.org/abs/2206.05235v1

110

Rachel Leah Childers @donskerclass.bsky.social · 16/07/2025

Galaxy brain take: these are all splines. Polynomial is a spline with 0 knots. Categorical and binning are 0th order splines of different bandwidths. Covariate adjustment, CIs, and order selection can be done as standard in GAMs. This is the philosophy behind binsreg: nppackages.github.io/binsreg/

Fig 4 from Cattaneo Crump Farrll Fung (2024) "On Binscatter"
https://maxhfarrell.com/research/Cattaneo-Crump-Farrell-Feng2024_AER.pdf

Panel A, B, C "Oversmoothed", "Optimal", and "Undersmoothed" show binned scatter plots with J=5, 11, and 50 categories, respectively, and accompanying best fit line.

1150

Rachel Leah Childers @donskerclass.bsky.social · 28/06/2025

Rakesh Vohra is pretty good at titles

Bayesian Bullshit
Sajan Srivastava Tymofiy Mylovanov Rakesh Vohra

August 2024

Abstract

A bullshitter neither knows nor cares about the truth, and therefore, it has been asserted, is more pernicious than a liar. We examine this assertion within the standard model of cheap talk communication where a bullshitter is modeled as an uninformed Sender. We show that in some circumstances, uncertainty about whether the Sender is informed or not can increase the welfare of the Receiver.

Keywords: Bayesian Persuasion, Information Design, Cheap Talk
JEL Classification: C6, D82, D83

140

Rachel Leah Childers @donskerclass.bsky.social · 23/05/2025

Casey Corgi Childers September 28, 2010-May 23, 2025 He was a very good boy.

Casey Corgi Childers, cuddling his pillow, earlier today.

7220

Rachel Leah Childers @donskerclass.bsky.social · 12/05/2025

The new theoretical aspects of M-estimation w/ ∞-dim nuisances are mostly notation: same estimator & formulas, though with a generalized notion of derivative, under regularity. The actual challenge is performing the optimization. Also calculating derivatives, but bootstrap avoids that.

Screenshot of tweets:

Quoting Tweet: I see that you have taken on a weighty vow. I respect it. tell the world.

Quoted tweet (by me): PSA that due to long-ago choice of username, I am morally obligated to engage with any tweet regarding Donsker's theorem. Please use this power wisely.

020

Rachel Leah Childers @donskerclass.bsky.social · 03/05/2025

As the semester winds down I have a gap before student final papers come in in which to read general interest economics. Anyway, Suresh Naidu, Winston Chen, & @treballen.bsky.social have fascinating work in progress on a quantitative GE spatial model of US slavery. dl.dropbox.com/s/uvhbdrqr88...

Estimated productivities, amenities, and coercion

Maps of the US South c1860 with county-level values from an estimated structural model based on population and production data. Of the 3, the most visually notable is the estimated coercion, a measure of the gap between enslaved and free worker populations, which concentrates in the cotton belt and around the Mississippi. Results in the presentation show amenities correlate negatively with malaria, while productivity shows weak correlation with cotton production, which should explain the lack of geographic concentration,

How does emancipation affect the allocation of labor?

Maps showing the equilibrium spatial allocation of formerly enslaved and free population in a counterfactual exercise in which slavery and coercion are completely removed in the US South in 1860 (probably not a fair comparison to actual emancipation due to continued coercion and restrictions on mobility post-bellum). According to this model formerly enslaved people would relocate from the cotton belt especially to Missouri, Appalachia, and Florida and formerly free people would make the reverse change. I imagine historians with regional knowledge would have plenty to say about the specifics.

010

Rachel Leah Childers @donskerclass.bsky.social · 01/05/2025

I don't know; I'm pretty happy to let non-specialists come to me and learn to understand what I know starting from first principles, with time and effort. Whether this counts as "accessible" might involve a discussion with our Office of Student Aid, though.

Table of Bowdoin 2024-2025 Cost of Attendance
Direct Costs
Tuition $67,198
Fees $634
Housing $9,148
Food $9,340
Direct Costs Total $86,320

Indirect Costs
Books, Course Materials, Supplies and Equipment $840
Miscellaneous Personal Expenses $1,660
Transportation varies
Indirect Costs Total $2,500+
Total Cost of Attendance $88,820+

This page is meant to be an overview and an introduction to costs and aid at Bowdoin. When you're ready to talk about detailed numbers and processes, the Office of Student Aid employs five of the best listeners we've ever met—they work with each applicant's family, one-on-one.

150

Rachel Leah Childers @donskerclass.bsky.social · 13/04/2025

Oh, thanks, I completely missed that! Expect more vibrant threads on Bayesian updating from here on out.

040

Rachel Leah Childers @donskerclass.bsky.social · 31/03/2025

There are some very unique suggestions about unconventional sources for a medical literature review in this episode.

Header for Archive of Our Own followed by search results:
"You searched for: back pain
sort by: best match descending
22,199 Found"

010

Rachel Leah Childers @donskerclass.bsky.social · 28/03/2025

Went to look up textbook results after getting the nagging feeling that an ML paper was reinventing classical ideas, and found this gem: "Not reading to the end of Le Cam's papers became not uncommon in later years. His ideas have been regularly rediscovered." At least they're in good company.

Text from van der Vaart, "Asymptotic Statistics" Ch 27, http://www.stat.yale.edu/~pollard/Books/LeCamFest/VanderVaart.pdf

The theorem may have looked to somewhat too complicated to gain popularity. Nevertheless Hájek's result, for general locally asymptotically normal models and general loss functions, is now considered the final result in this direction, Hájek wrote:

"The proof that local asymptotic minimax implies local asymptotic admissibility was first given by LeCam (1953, Theorem 14). ... Apparently not many people have studied Le Cam's paper so far as to read this very last theorem, and the present author is indebted to Professor LeCam for giving him the reference"

Not reading to the end of Le Cam's papers became not uncommon in later years. His ideas have been regularly rediscovered

1162

Rachel Leah Childers @donskerclass.bsky.social · 26/03/2025

Congrats on reinventing the analysis and synthesis filter banks for the Discrete Wavelet Transform.

Iterated analysis filter bank, with 2X downsampling

Diagram from https://rafat.github.io/sites/wavebook/intro/fb.html

Iterated synthesis filter bank, with 2x upsampling

Taken from https://rafat.github.io/sites/wavebook/intro/fb.html

010

Rachel Leah Childers @donskerclass.bsky.social · 19/03/2025

Cute idea. Imposing additional regularity on the difference in means has been shown to improve minimax rates in @edwardhkennedy.bsky.social et al arxiv.org/abs/2203.00837 not in an RD setting or for inference. It makes me think similar techniques might apply beyond the exactly linear CATE case.

Figure 2 of https://arxiv.org/abs/2203.00837

Graph showing convergence rate for CATE estimation as function of smoothness of conditional means and the CATE function itself (Hölder exponents "s" and "γ"), showing that smoothness of the CATE determines the rate at the typical nonparametric rate if the means are sufficiently smooth (after the "elbow") while both matter if the means are sufficiently rough.

Caption: "The minimax rate for CATE estimation, as a function of average nuisance smoothness s and CATE smoothness γ, each scaled by covariate dimension d. The black dotted line denotes a threshold on the nuisance smoothness s/d, below which the oracle nonparametric regression rate n�1/(2+d/γ) is unachievable (the “elbow” phenomenon)."

171

Rachel Leah Childers @donskerclass.bsky.social · 07/03/2025

I made it official! I am now legally Rachel Leah Childers.

State of Maine
Cumberland County Probate Court

In Re: ***** ******** CHILDERS CERTIFICATE OF NAME CHANGE

I, Susan G. Schwartz, Register of Probate of Cumberland County Probate Court in and for said County, on the 7th day of MARCH 2025 the name of ***** ******** CHILDERS of BRUNSWICK, ME in said County was changed to RACHEL LEAH CHILDERS which shall hereafter be this person's legal name.

IN WITNESS WHEREOF, I have hereunto set my hand and affixed the official seal of said Court, this 7th day of MARCH 2025.

S. G. Shwartz
Register of Probate

61201