I've recently started going through Axler carefully and doing the problems, a quarantine activity I guess, and have been enjoying it. I actually learned about this book on an older HN post.
It does have plenty of matrices. The main thing it really does is avoid determinants until the very end. The determinant is certainly something I remember learning as a kind of rote operation, without really understanding any intuition behind why you'd multiply and add these numbers in this particular way. I still feel lacking in "feel" here, which is why I suppose I'm going through Axler now.
The asset value underlying an ETF can be negative, but then the owner of the ETF will just have a worthless piece of paper. The ETF managers will have a problem on their hands regarding the negative amount, though.
In contrast, a futures contract is an agreement to make a future trade, so it can keep going against you past 0. If you are able to take physical delivery, your worst-case scenario is that you pay the money you said you would, and you get your oil. But if you are a casual day-trader type, you probably don't have the ability to take physical, so you may be in trouble (over a barrel, literally).
I agree with you about the dangers of ETFs and about knowledgeability. I didn't mean to advocate for ETFs on an absolute basis, just to make a relative statement about them vs. futures.
I'm not recommending USO or oil itself as an investment, and I agree that you will be paying money to the ETF manager if you get involved in it.
But I don't agree that USO is _more_ dangerous for a casual trader to trade than oil futures, for the reasons I mentioned.
Removing the overall fluctuations of the oil market, the relative problem with ETFs is that they bleed away value over time. That's a real issue, but I wouldn't compare that to juggling a live hand grenade.
Edit: you did not say it was more dangerous, my mistake. I do think that ETFs are less dangerous, for the reasons I mentioned.
There are ETFs that track oil futures (basically like a stock, but backed by oil instead of a company). It's been a while since I've looked at any of this, but I think USO is still the most prominent.
There are plenty of things to watch out for with these ETFs. You pay ongoing expense fees. And ETFs, especially those that aren't just holding containers for assets, can have subtleties in their prospectuses that cause their value to fluctuate in counterintuitive ways. There's still a lot to be cautious about.
However, compared to the actual futures, they're more suitable for casual investors, for reasons such as what we see here. They can't go below 0 and don't necessarily involve margin. And the ETF will typically deal with things like rolling the futures position ahead of expiry.
I don't know that this is regional so much as colloquial vs. formal. In the US people recognize that the subjunctive is technically correct and use it in formal writing. But it's very common to violate this rule in speech, so much so that to my ears it sounds less "wrong" than "informal."
As far as musicians go, you can find examples of people on both sides of the pond using or not using the subjunctive. Beyoncé sings "If I were a boy," while Thom Yorke sings "I wish I was special."
Based on my experience learning Czech (not native at all, just interested):
- it's typically listed as a separate letter when writing out the alphabet
- but in practice it's typed out as "c h" and not as a single character
- it occupies its own place in Czech standard alphabetical order, my English-Czech dictionary has all the "ch" words after "h" (so interestingly in order to do a proper sort programmatically you need to possibly look 2 characters ahead)
Since last Friday I've been making my Zoom virtual background a daily rotation of artwork. It's been a fun and educational experience for me (I'm no art expert, just a casual fan of museums) and I think my friends and co-workers have enjoyed it. When you're mostly stuck at home all day it's nice to have a daily "new" to look forward to.
I've been recording my choices and when this is over I'll have a little "Zoom Museum" as a kind of artistic memorial of the pandemic.
This link is a great source for new art, so thank you OP!
I think the part that you have correctly included that people forget or elide is that it's the probability under a specific null hypothesis. So it's a function of what you have chosen for that - normal distributions, a certain parameter value of 0, etc. So this means that a) it's not the probability you'd see in the real world under repeated performance b) it's not the probability under other reasonable null hypotheses. Like maybe under the hypothesis parameter = 0 you get an improbably large p-value, but under parameter = 0.1, or with different assumed underlying distributions, you wouldn't see something so extreme.
I'd guess that the original writer understands this, and that Gelman is only pointing it out because casual readers sometimes don't mentally retain the full baggage that the p-value carries.
Watterson highlighted that arc in his 10th anniversary collection (which is a really wonderful collection of essays and explanations side by side with strips). It happened relatively early into the comic strip's life and IIRC he regarded it as a turning point in Calvin & Hobbes's ability to move beyond gags and handle weightier topics.
There are a lot of crossword constructors who publish/sell their puzzles independently on the Internet, and often they create both puz and pdf files. Some sources of free ones:
It would be interesting to see how realistic a cat thinks these are, maybe by measuring brain activity or reactions. It's possible that a cat may not be fooled by cats we think look real, or perhaps more interestingly, that a cat is fooled by a not particularly good image.
Common cuckoos lay their eggs in other birds' nests. The chicks don't necessarily look much like the host species to the human eye, but they can fool their hosts along the correct dimensions to get food from them. It's an interesting question to what degree ML algorithms trained on human dimensions could be foiled by an animal whose brain has been wired for different perceptions, or how feasible it is to train an ML algorithm on animal perception, or if it's possible to make an algorithm that successfully fools, say, both man and dog.
On the last point, for example: to make fake sounds that fool animals with different hearing ranges, presumably you have to be able to output sounds across the union of the ranges and train on sound data over the union of the ranges.
(Note: I'm not a biologist, if someone more informed wants to correct me on anything here you are welcome to do so.)
> Spend 50% of the writing time actually writing, the rest tweaking, reading and illustrating. Details are important.
I find this true not just of blog posts, but any time I've done formal writing at all. There is an immense gulf in quality between writing that has gone through even just one heavy edit/revision process and writing that never has.
I've recently been encouraging my technical co-workers to expend more time on not-code, such as documentation, tutorials, plans, postmortems, internal RFCs, carefully worded PR/commit comments, etc. One challenge is that people sometimes think that when they "finish" a piece of writing, they're done. Unless you've been iteratively revising pieces of it in smaller chunks, I think you're really only halfway there.
I haven't made the donation yet for 2018, but I make donations every year right at year's end to at least these two charities:
- Doctors Without Borders
- Direct Relief
Global health problems is where I'm most interested in donating money. I split my money between those two primarily to have a little diversification from idiosyncratic risks to a single operation.
These occupy the vast majority of my charitable giving. I give or have given in the past much smaller amounts to a smattering of other charities: Wikimedia, Give Directly, ACLU, Planned Parenthood, Lambda Legal, Girls Who Code.
Also: I'm an avid crossword solver, and over the last year or two some puzzle constructors, as a way of doing good, have started offering packets of crosswords in exchange for donations to a cause. The original, I believe, was Francis Heaney's Puzzles for Progress http://puzzlesforprogress.francisheaney.com/ for broadly progressive causes, but I have also seen Queer Qrosswords https://queerqrosswords.com/ for LGBTQ charities and Women of Letters https://www.pattivarol.com/women-of-letters/ for feminist charities. If it tips you over the edge to donate, do it and go get some puzzles!
- _SPQR_, by Mary Beard. Engaging book surveying the history of ancient Rome, mostly Republic and early Empire if I recall correctly.
- _To Explain The World_, by Steven Weinberg. History of physics from the ancients to about the time of Newton. Don't skip the technical notes! Actually do the problems!
- I reread _Wolf Hall_, by Hilary Mantel, it was as good as I remember. This time through, I also spent some time on the Internet tracing the histories of the major characters before and after the events of the book, and it really enhanced my appreciation of it. (I also read, for the first time, its sequel, which was fine but not quite as good.)
Yes! Although I was an econ major I did the same thing, I dipped my toes into pure math and took some real analysis and abstract algebra. I really found these challenging at times. But the ideas and modes of thinking I gained from these classes have been very rewarding since.
I relate its value in programming to the Torvalds adage "Bad programmers worry about the code. Good programmers worry about data structures and their relationships." Taking some abstract algebra really helps you think about data structures and their relationships, and to architect "good bones" in your code.
Yeah, I'd be very interested to learn about anything about the sensitivities of ridership to other external variables in general.
Looking at the article's linked presentation [1] and the MTA's most recent financial plan [2] it seems like all the hurt is coming from really huge declines in projected revenues - labor costs seem to be growing pretty reasonably but there's basically zero projected growth in fares. If someone can give me a layman's explanation of what "Capital and Other Reimbursements" is, which accounts for about a $500mm decline between 2019 and 20222, I'd be much obliged.
So I am curious how sensitive riders are to the increased service problems, how much that makes them switch out, to get some sense as to how much the signal improvements will help solve this problem. Also how much to a fare hike, which seems like the more straightforward answer in a vacuum (i.e. other than taxes or other government infusions). The MTA says in [1] that even "draconian service reductions would have a relatively small impact on the deficit."
I don't know anything about the Israel-Intel deal and have not decided personally how I feel about HQ2 in NYC. But I upvoted your comment because it emphasizes looking at other historical deals, and not just the projections and analysis on the current deal, and would love to read more of that in this discussion. This is a very large project over long horizons (10yrs) and it's hard to imagine being anything but extremely uncertain about projections; you have to look at history.
I'm also a sports fan, and I've read about the dismal returns to taxpayer-subsidized new stadiums despite rosy promises of economic development, and tend to look askance at this kind of thing.
I really like your last point "whatever the computer is telling is true from the computer point of view".
In a surprising number of coding interviews that I've conducted, I've seen a candidate write an incorrect solution that generates an exception and then attempt to figure out the error almost on first principles, rather than actually reading the exception carefully and thinking about what coding errors could have caused it to happen. The exception message is a _huge_ hint!
It's a red flag to see someone try to debug without paying attention to it. Conversely it's very positive to see someone encounter one and then think carefully about it. Writing an incorrect solution but then testing it out and showing the ability to debug and fix it systematically and expediently is no worse than getting it right on the first try, in my book.
I like that Victor Hugo's 19th century version of "deleting apps from your phone" is "deleting your clothes".
I have found that an easy, milder nudge, for either computer or mobile device, is to remain logged off of things by default and explicitly log out whenever you're done.
It does have plenty of matrices. The main thing it really does is avoid determinants until the very end. The determinant is certainly something I remember learning as a kind of rote operation, without really understanding any intuition behind why you'd multiply and add these numbers in this particular way. I still feel lacking in "feel" here, which is why I suppose I'm going through Axler now.