huitzitziltzin

49 ms·

481.
by huitzitziltzin 6y ago
> I personally think a lot of established macroeconomics is bullshit. There's just no mechanism to verify many of the hypotheses Modern academic (and central bank) macroeconomics is literally all about taking macroeconomic models to
482.
by huitzitziltzin 6y ago
This is not a good recommendation. Wray and the MMT school have a bizarre set of views which are wildly outside the mainstream of macroeconomic thinking. Moreover, when serious macroeconomists have tried to engage with MMT on its own term
483.
by huitzitziltzin 6y ago
> don't think this is exactly malicious, but it bothers me that the field largely operates on unchallenged assumptions What do you know about the research done by contemporary macroeconomists? Anything? Do you know how it is tau
484.
by huitzitziltzin 6y ago
> Economists are in broad agreement because they're paid to be. This is well-documented historically. This is *complete* nonsense. No one pays me to think or say anything. No one pays anyone in my department anything to take a par
485.
by huitzitziltzin 6y ago
People claiming that Apple can do "whatever it wants" are definitely wrong, but in all of the discussion on this article I don't see anything like the following: which current antitrust law is Apple violating with this beha
486.
by huitzitziltzin 6y ago
> after controlling for physical activity You say that like it's easy to do. It's not. That's not directly measurable, certainly not across generations. It's only barely become measurable in the past five or six ye
487.
by huitzitziltzin 6y ago
>We don’t know why. If I were to guesss, due to the use of plastics with our food. What evidence do you have to dismiss the (IMO much, much more plausible) idea that it's just men becoming more sedentary, moving and exercising less?
488.
by huitzitziltzin 6y ago
Where do you get that from? The FTC has challenged a lot of hospital mergers. They are not always successful, but they do challenge them. Source: healthcare/competition economist.
489.
by huitzitziltzin 6y ago
Kaiser might be a weird outlier. IT is a very different kind of organization (vertically integrated - physicians/insurers/hospitals). It might be that the price is that insane elsewhere, but Kaiser may not be the best way to judg
490.
by huitzitziltzin 6y ago
AFAIK, certain non-profits are required to report the salaries of their ten highest paid employees. Doesn't mean it will be easy to find, but you might be able to find it that way. But this is not information the hospital would share
491.
by huitzitziltzin 6y ago
Everyone hates pharma (not entirely without reason!) but all drug spending is about 10% (slightly less) of US health care spending. If you spent $0 on drugs next year, we would not save that much money. If you want to cut spending, you
492.
by huitzitziltzin 6y ago
Underrated answer, but it isn't an anti-trust violation when they have permission to accredit med schools! To be clear: I totally agree w/ you that the AMA is a cartel.
493.
by huitzitziltzin 6y ago
Do not draw conclusions from this specific case because Kaiser in particular is very, very different. You may be able to find similarly insane examples at other hospitals, but Kaiser is vertically integrated: generally Kaiser doctors send
494.
by huitzitziltzin 6y ago
Personal opinion (as a health economist) is that this is a nice policy, but probably not going to do that much. You as a consumer now have the ability to search, but do you have an incentive to do so? Do the savings come to you? Some of it
495.
by huitzitziltzin 6y ago
The Health Care Cost Institute has something like what you want - see Figure 2 about halfway down this page: https://healthcostinstitute.org/hcci-research/international-...
496.
by huitzitziltzin 6y ago
>pay for US doctors makes up a fairly small portion of overall US medical expenditures (less than 10%) That is likely too low. The Centers for Medicare and Medicaid Services provides a National Health Expenditure estimate annually. [1]
497.
by huitzitziltzin 6y ago
> The simple answer is that Americans really do consume significantly more healthcare than Europeans. I do not believe this is correct. From the OECD, average length of hospital stay across rich countries: https://data.oecd.or
498.
by huitzitziltzin 6y ago
This is not that plausible as an analysis of the role of chargemasters. I agree that the chargemaster prices are fake, but what you are proposing is: Hospital: "Our insurance partners pay us a fraction x of Chargemaster charge X, so l
499.
by huitzitziltzin 6y ago
IMO, array indices are a matter of taste which inspire feelings of religious intensity. You can change it if you prefer 0 or -14 or whatever number you like. Non zero effort is required, but if you otherwise like the language it can be cha
500.
by huitzitziltzin 6y ago
Not sure about your exact use case and this is not part of Julia which I have ever used myself but the answer might be here: https://github.com/JuliaLang/PackageCompiler.jl There is a presentation from a recent JuliaCo
501.
by huitzitziltzin 6y ago
Maybe keep an eye on this issue: https://github.com/FluxML/Flux.jl/issues/1431 They are going for feature parity with pytorch hopefully in the near term.
502.
by huitzitziltzin 6y ago
>House prices aren't included and we've seen prices balloon over the last decade Depending on exactly what you mean, this is not a great criticism of the CPI. House prices are excluded on purpose. Housing costs are include
503.
by huitzitziltzin 6y ago
The first author (MK) also wrote "Algorithms for Optimization" (MIT Press), which is also implemented in (and is a nice guide to) the Julia language, which is itself a pretty sweet language.
504.
by huitzitziltzin 6y ago
Two separate issues here - 1. Make the underlying model public (e.g., you ran such and such regression, here’s what you included and here’s a table of coefficients). Inarguably a good idea. Totally unobjectionable. You should correct
505.
by huitzitziltzin 6y ago
These files are more widely available than people think. For California for example, an agency called OSHPD makes available a set of documents called “charge masters” which have similar information. This appears to be a little different bc
506.
by huitzitziltzin 6y ago
Covered in "Dive Into Deep Learning," which another commenter has recommended. https://d2l.ai/chapter_attention-mechanisms/transformer.html
507.
by huitzitziltzin 6y ago
I really like Aggarwal's "Neural Networks and Deep Learning". I recommend it highly. (At least this summer Springer was giving away a PDF - not sure if that's still true.) To me, Goodfellow et al. spent the first hundr
508.
by huitzitziltzin 6y ago
Here's a broader set of benchmarks, including to R, Matlab and Fortran: https://julialang.org/benchmarks/ AFAIK these have been optimized for the other languages, but your opinion may differ if you look at the imp