hnracer

.

Are you saying the book itself is ML porn?

Indeed but I wonder if one could build a platform that does this behind the scenes. User sends BTC to the online wallet, and the site hedges on behalf of the user on one of the liquid future exchanges, locking in the prevailing BTC price. People can make payments out of this online wallet with just as much ease as they're currently doing. Then they have an easy to use digital cash without any price risk.

The site charges a small fee for the service of derisking the holdings.

This does seem to solve a real user pain point (risk and the complexity of hedging) and benefits from scale economies.

You can just hedge your exposure on a futures exchange and still use BTC to transact but be unexposed to any price movements. Although doing that is very complicated and defeats the purpose of an easy to use alternative to centralized cash.

How do they account for the expected lower turnover due to the price increase when they estimate the post-policy gross hourly wage?

It's not an ideal natural experiment because the cost of all ride sharing will rise simultaneously, so many of the substitutes (other brands) are equally more expensive. In other markets Uber couldn't raise their prices without having their market share decimated by Lyft, unless there's collusion.

I think there's two reasons.

First is that the costs are hidden and complicated. We need a decent understanding of economics combined with an honest and deep analysis of the consequences of the specific policy to conceptualize why it's deleterious. Very few people will tick both boxes.

Second is that politicians want to be seen to be doing something. Even if leaving the market alone is the best of all options.

Does R have robust CSV parsing? I remember using the default and it'd be extremely finicky about getting the header and index flags right and wouldn't typecast numeric columns properly (instead they'd end up as factors and not play nice)

I've used R (3 years) and Python (8+ years) in data science and much prefer Python, because it can do things that aren't just pure data analysis, and because pandas is so amazingly good compared to R's data matrix solutions, in my opinion. I believe that the algorithmic trading industry has gone fully into Python and away from R for these reasons.

It's designed to be a 1v1 game, a lot of people don't like that because it becomes hyper-competitive and even a little stressful when compared to team games

Is there a good discovery mechanism for finding individual journalists producing quality content on eg substack or medium? I want to find their content but I don't know where to start.

The burden of proof is on the individual claiming that agricultural investment by Western firms into Africa is a negative, given that it's commonly believed that FDI is a good thing and African governments routinely and actively seek out FDI as part of their policy. So I don't believe that to be question begging.

The sentence following the one that you quoted is the defense, perhaps you should argue against that instead?

RTSs attract a certain personality type: the type of person that becomes an engineer, mathematician, scientist, or statistician. I met so many individuals from these career choices when I was playing RTSs that it was uncanny. Given this, it's no surprise to me that RTSs aren't wildly popular, the complexity and depth is just too much, it takes a systematizing nerd (of which there aren't many) to fully appreciate it.

I believe another reason is that these games are mostly 1-on-1, so they don't benefit from the same team play benefits and network effects as other games.

We can't honestly equate the Uyghur genocide, jailing of political dissidents, Xi's dictator for life status and the CCP's censorship apparatus to the modern US, even when factoring in the ongoing travesties e.g NSA surveillance, unequal policing, and I could even throw in Epstein's killing in there if it makes you feel better.

These are not in the same ethical ballpark, and one doesn't have to be a US nationalist to recognize this.

Apologies, I meant to say large public companies in my OP, i.e. those listed on the stock exchange. I'd be interested if there was any examples of those using compromising political information for leverage. I understand however that if it happened we may not have heard about it.

It's the proximity of the situation to Jack Ma's comments that tips you off as to the true reason for the actions. It's not just normal anti-trust, the timing suggests an attempt to punish any opposition to the CCP or Xi.

And that's not a small difference, it's the difference between a liberal democracy (with its many mistakes and flaws) and an authoritarian dictatorship. The both-sides fallacy should be avoided here and elsewhere.

It's wild and rather irresponsible given the censorship and surveillance that people in China are subjected to.

Are there examples of large public companies releasing compromising information on elected officials in response to political threats or pressure?

#84 is especially good advice. Bandwagon criticism and demonisation is very common in the workplace and pretty much every group environment I've been in, although to varying degrees. Definitely a deep-seated psychology/biological phenomenon at work.

I'm probably being dense but what do you mean by "the same way it was when it was tossed 51% of the time"?

There's two problems in markets, price prediction and execution (ie what to do with your prediction). The former is a supervised learning problem but the latter is an action space problem ie an RL problem. Although nobody in industry has gotten any RL methods too work, they overfit to the incredibly small data sets.

Many very smart people have tried and failed. State of the art remains very basic supervised models with hand engineered features. In the markets, data is permanently scarce, so these methods don't work well. In the RL problems that DeepMind is solving, data is literally unlimited, and that's the problem space that these methods have been designed for.

There's ways to address this such as an open auction with one day to collect buys and sells. You'll get a pretty stable open price determined by the fair open market.

I share your view. My strategy in university was to memorize as much as I could going into the exam and then garbage collect 1 day later and be left with nothing.

This was an arrogant waste of time, but my priorities at that time were to simultaneously max(grades) and min(effort), with the end goal being a job, and my strategy optimally satisfied those goals. In hindsight I wish that things had been designed differently to prevent gaming the system in such a way, to save me from myself, but alas they were not (and still aren't).

At least in my cohort and my university, cheating was endemic and done by almost everyone on every assignment. Not cheating was a significant disadvantage. I didn't ever hear of anyone getting caught which I find astonishing, since I'm sure the graders were faced with multiple identical answers.

Maybe because the subjects were quantitative, there was less scope for plagiarism detection? Or maybe they just didn't really care, or they didn't want the administrative trouble. I'm not sure.

I think we are mostly in agreement. I agree that there are societal factors as well and that these are not good and we should work to address them.

It's good to keep in mind that not all medical services are price inelastic in the sense that ambulance costs or emergency hospital visits are.

GP visits/non-emergency general surgeries/general medicines - just like food, are all necessities for survival, and are all locally price elastic, with the exception of medicines that have regulatory capture - not a free market, such as insulin which is many X the cost in the US because of a government-enforced oligopoly.

This means that no single supplier (whether of food or non-emergency medicine) can raise their prices because consumers have time to find a cheaper alternative. Unless they collude, which is illegal, and impractical for many markets such as GP visits.

For these particular goods, does your view differ, when compared to emergency services which we agree are price inelastic? And if not, how do such services differ to the foods market - holding aside regulatory capture of things such as insulin which we both probably agree is not a good thing?

Agree that this is an issue