stdbrouw

43 ms·

421.
by stdbrouw 10y ago
To avoid adding to the misinformation and ignorance out there, what are some of these edge cases and minor details?
422.
by stdbrouw 10y ago
The VAT for all costs you incur can be deducted, regardless of whether they're fixed or variable. If you're operating at a loss or if you do a lot of business with places where you don't have to charge VAT, it's not impo
423.
by stdbrouw 10y ago
> if they have any EU customers they must either VAT register in the UK (which adds 20% to their UK prices) While I'm sympathetic to the fact that this will increase your administrative burdens, in terms of your bottom line this rea
424.
by stdbrouw 10y ago
It's not easy by any means but neither would I call it a huge, overcomplicated mistake. Even the new rules for digital services are overdramatized – it's really just as simple as charging a different VAT rate per country and submi
425.
by stdbrouw 10y ago
The Hawthorne effect describes how any change in environment can temporarily lead to positive effects. The sentence you quote is more akin to Goodhart's law ( https://en.wikipedia.org/wiki/Goodhart%27s_law ).
426.
by stdbrouw 10y ago
You might still wonder whether, insofar as modern society works due to a sort of non-aggression pact, all of us are getting the same benefits from that pact.
427.
by stdbrouw 10y ago
The author wonders whether there will come a day when see Mars as not just something we exploit but as a home we must protect, just like Earth.
428.
by stdbrouw 10y ago
> Alcohol would have the advantage over boiling water that the bacteria would not be in the alcohol in the first place so they could not have produced something bad for you in the water. This might hold for liquor (depending on what dest
429.
by stdbrouw 10y ago
This has to be the worst imaginable example. Have you never seen a coin-operated vending machine?
430.
by stdbrouw 10y ago
It's known as belief perseverance ( https://en.wikipedia.org/wiki/Belief_perseverance ). A well-studied recent example is how people changed their beliefs after it became known that Iraq did not have weapons of mass
431.
by stdbrouw 10y ago
But now you've gone from "huge opportunity" to "maybe something that can be used for some pharmaceutical products".
432.
by stdbrouw 10y ago
Given that the alternative hypothesis is usually defined as the complement of the null hypothesis (HA = 1 - H0) this doesn't make much of a difference, though.
433.
by stdbrouw 10y ago
An example from Leonard Mlodinow: imagine a friend doesn't pick up the phone when you call. Now, you might think that maybe they're upset at you, because if they are indeed upset, the the probability that they wouldn't pick u
434.
by stdbrouw 10y ago
But that's the thing, there is just never any real need to do hypothesis testing, the range of plausible effect sizes is all that matters.
435.
by stdbrouw 10y ago
> If clinical trialists use p-values wrong, how is moving to Bayesian methods going to be less misused and misunderstood? Bayesian methods have the advantage that, while they can still be misused and misunderstood, at least when used cor
436.
by stdbrouw 10y ago
> Frequentist and Bayesian statistics are different tools that answer different questions. Arguing whether one is better than the other is like arguing whether one should drink water or eat bread. Astronomy and astrology also answer diff
437.
by stdbrouw 10y ago
> A good book that talk about this which I've been meaning to read is... In All Likelihood: Statistical Modelling and Inference Using Likelihood by Yudi Pawitan. It's good but it's mostly a technical introduction to statis
438.
by stdbrouw 10y ago
> Really, my blog kind of missed the mark. It'd be sad if people only published stuff once they considered it perfect. I enjoyed the post, thanks for writing.
439.
by stdbrouw 10y ago
... and in fact, it's machine learning models whose performance is tested using cross-validation or a holdout sample that assume new points are drawn from exactly the same distribution that generated the training data, whereas Bayesian
440.
by stdbrouw 10y ago
Initially you mentioned that 47% is fine, now <33% utilization is also fine, okay, but what kind of utilization isn't?
441.
by stdbrouw 10y ago
The article doesn't address this, but the interactive graphic does: 831 out of 1170 California power plants (71%) are operating at less than 1/3 of capacity.
442.
by stdbrouw 10y ago
Not saying the article is perfect, but I'm seeing the same one or two counterarguments presented over and over again in almost identical fashion. Am I paranoid or are we being astroturfed?
443.
by stdbrouw 10y ago
If you've got a bunch of ordered categories, like "bad", "okay" and "good", the category in the middle. (So if humans can have 0, 1 or 2 legs, then 1 would be at the center.)
444.
by stdbrouw 10y ago
Think Stats ( http://greenteapress.com/wp/think-stats-2e/ ) and the follow-up Think Bayes are awesome if you're a coder. https://bblais.github.io/statistical-inference-for-everyone-... is a com
445.
by stdbrouw 10y ago
Lauded, sure, but just about all of his books are out of print and his work is rarely a part of the curriculum.
446.
by stdbrouw 10y ago
> In general, the median conveys remarkably little information. It's great at showing the center, but it gives zero indication of anything else. The median also has zero ability to make me coffee in the mornings, but I don't th
447.
by stdbrouw 10y ago
> it is pretty amazing that normit transformations (map the quantiles of a non-normal distribution onto a Gaussian and use that) don't seem to be on this guy's radar I don't mean to offend, but this is the PhD ur-response,
448.
by stdbrouw 10y ago
> OK, so what else do you suggest? Slopes, a.k.a. the parameters in a regression analysis, ideally as a confidence interval or prediction interval to account for uncertainty. Interpreting and communicating regression analyses is a pretty
449.
by stdbrouw 10y ago
Something like what you describe is more or less how statisticians do in fact reason about statistics: https://en.wikipedia.org/wiki/Sufficient_statistic#Minimal_s... , though only for i.i.d. data from a parametric dist
450.
Unlearning descriptive statistics(debrouwere.org)
343 points by stdbrouw 10y ago|92 comments

More ›