81–90 of 96 posts
Earlier quoted context omitted.
This seems like a very good use case for GPT models: Surfacing buried information.
I've had this thought before -- that there's lots of good ideas and insights hidden in the deluge of comments on hn, reddit etc -- and GPT might be used to find them. But I haven't figured out a good prompt yet.
Earlier quoted context omitted.
I've had this thought before -- that there's lots of good ideas and insights hidden in the deluge of comments on hn, reddit etc -- and GPT might be used to find them. But I haven't figured out a good prompt yet.
This is more of a data question rather than finding a prompt.
Earlier quoted context omitted.
SEC’s EDGAR database has the filings available to everyone the moment they’re released. It will take you 5-10x longer to press F5 than it takes a bank/hedge fund to have a bot ingest the news and trade off of the new information. Even if you could ingest the information instantly, your orders will be slower than the people who are co-located within the same data center as the exchange, or the folks who set up private…
The very fact that people go to such lengths to rig trading systems in their favor suggests they're not worth retaining in their current form.
Earlier quoted context omitted.
Before you think about making something that quickly reads 8-K filings and acts upon it: don’t bother. A bunch of entities are already doing this way faster than you ever can through various methods and by the time you get the data the opportunity to act is gone.
Nasdaq closing auction ends at 4pm. Reports come out later. Participants trading on 8-K cannot trade the news immediately. Orders can be executed only on Monday during opening auction. Also, you assume their trades are enough to move the market and neutralize the opportunity, which may not be the case. Perhaps they can trade off market. I’m just saying that it may not be a done deal as you describe it, but I may be m…
Earlier quoted context omitted.
Never. We are humans - humans are by our very nature dynamic.
You mean we have to always oppose someone else’s incentives? Or we always have to see someone else’s incentives as an opposition to ours? What does dynamic nature imply? Can’t a system be dynamic?
Systems can be dynamic but humans are more creative dynamic than systems can be designed. That is and will always be true.
Earlier quoted context omitted.
> I have idly wondered whether there's an viable opportunity to exploit retail investors who're being manipulated/misled (by, e.g., MSM financial 'news', stock-picker personalities, bloggers, WSB, etc.). Generally, this idea doesn't work because "manipulated"/misled retail investors will still be right 50% of the time. Being bad at trading gets you zero EV, not negative. But if you look closely, those retail investor…
> Being bad at trading gets you zero EV, not negative. I’d have to think a bit more to be sure, but my first instinct is that “bad at trading” would be negative EV given the presence of firms who are good at it. If you enter a trade in the “wrong direction”, you’re much more likely to get a fill than if you are in the right direction. Other errors include letting losers run (hoping to “get out even”) and cutting winn…
Takers (ie. most retail investors) always fill. Makers create an order book, defining a buy (say $101) and a sell price (say $99), along with a spread in-between ($101-$99 = $2). If you buy a stock at a price the maker offered, you will be guaranteed to get it; the maker has no say in that transaction (they can't reject it anymore after the taker accepted).
If you do the math carefully, you'll see that both strategies "letting losers run" and "cutting winners short" fascinatingly have a neutral EV (ignoring fees and spread). You just take a lot of unnecessary risk/variance, but it all averages to zero. (Even the strategy "I just go all-in until I am bankrupt" has a zero EV after any finite number of iterations, because the exponentially unlikely chance of you winning every single time comes with exponential payoff.)
The fundamental principle behind this is that takers buy at market prices (ignoring fees and spread), and that those market prices are in an equilibrium. If it were possible to easily lose money by buying at market prices, then billion-dollar hedge funds would've already shifted the market prices by simply doing the reverse, before the retail trader would have any chance to trade on it.
Earlier quoted context omitted.
> Being bad at trading gets you zero EV, not negative. I’d have to think a bit more to be sure, but my first instinct is that “bad at trading” would be negative EV given the presence of firms who are good at it. If you enter a trade in the “wrong direction”, you’re much more likely to get a fill than if you are in the right direction. Other errors include letting losers run (hoping to “get out even”) and cutting winn…
> If you enter a trade in the “wrong direction”, you’re much more likely to get a fill than if you are in the right direction. Takers (ie. most retail investors) always fill. Makers create an order book, defining a buy (say $101) and a sell price (say $99), along with a spread in-between ($101-$99 = $2). If you buy a stock at a price the maker offered, you will be guaranteed to get it; the maker has no say in that tr…
Earlier quoted context omitted.
> If you enter a trade in the “wrong direction”, you’re much more likely to get a fill than if you are in the right direction. Takers (ie. most retail investors) always fill. Makers create an order book, defining a buy (say $101) and a sell price (say $99), along with a spread in-between ($101-$99 = $2). If you buy a stock at a price the maker offered, you will be guaranteed to get it; the maker has no say in that tr…
I'm a retail investor. I think I've entered fewer than 2% of my orders as market orders. I rarely want to take a price and give up the full spread. I doubt I'm alone.
Earlier quoted context omitted.
Your links don't reveal growing complaints , they're just complaints people posted to Reddit. Of course there will be tons of complaints, Amazon is huge. Picking out a few of them is meaningless.
> Your links don't reveal growing complaints, Actually, they do. People didn't complain about this really 2 years ago. For the past few months, its all over the place (social media: Reddit, Twitter, Tiktok, Instagram; OSHA fines; SEC anti-trust lawsuits). > Of course there will be tons of complaints, Amazon is huge. Picking out a few of them is meaningless. No, not really meaningless. It's indicative of the illegal p…
Companies (e.g., Amazon) hide negative news[1] with unrelated press[2][3] alongside SEC filings[4].
[1] See my original comment: https://news.ycombinator.com/item?id=35877291#35881269
[2] Amazon's new service lets you buy physical items in games and apps: https://news.ycombinator.com/item?id=35880117
[3] Prime Video's architectural improvements to Quality Analysis service saves $$: https://www.primevideotech.com/video-streaming/scaling-up-th...
[4] Amazon CEO of Retail is dumping stock: https://www.benzinga.com/sec/insider-trades/search/index?com...
Earlier quoted context omitted.
Thanks for the links. Interestingly, I noticed Amazon's CEO of Retail, Doug Herrington, dumping sizable amounts of Amazon stock[1]. This activity is coupled with growing complaints about Amazon's quality of goods[2], delivery[3][6] and ability to return those goods[2][4][5][8][9]. Amazon is now requiring customers to file a police report in order for their credible return to be accepted[7][10][11]. Amazon is also can…
When you notice that X person is "dumping sizable amounts of stock", and use that as your foundation for a wave of further speculation, the first thing you should consider doing is looking at the actual filing[0]. Which of course contains the pertinent information: "This transaction was effected pursuant to a Rule 10b5-1 trading plan adopted by the reporting person on 11/23/2022." So did this person "dump" stock? Sur…
...I did.
> So did this person "dump" stock? Sure. Did they choose to do it at this particular point in time? No, a predetermined trading plan did.
Doug began offloading stock 11/23/2022. Doug is still selling this month. While I agree that some was predetermined, it wasn't all predetermined. And that's the point.