cameronfen

Yes, if they're causing harm, certainly. btw, I don't know how accurate these types of studies are and what the actual ratio is, but I wouldn't be surprised if the percentage was higher than in the general population. Not all psychopaths are antisocial. Interesting read: https://www.smithsonianmag.com/science-nature/the-neuroscientist-who-discovered-he-was-a-psychopath-180947814/ I'm of the belief that having a small proportion of people with psychopathy or even narcissism disorder (although I can't think of a justification for that but I'm sure there is one), is beneficial to society. For example, I think most normal people would feel very hesitant to exploit third world countries with sweatshops for cheap labor, but if no one does stuff like that, there is never any investment in those countries as that is the first step to an educated workforce and a country with infrastructure to produce more advanced products.

Same idea as you with FB. Bought after CC, will reduce around 195 to my normal position.

haha thanks. Just to clarify though, I dont think academics are conciously cheating. But when you are pushed to get positive results, you typically search until you find something, then you just don't check to make sure you didnt screw something up.

not going to quote again but I took a look at their paper. You are certainly right the results seem to be robust to the param choice they show. That being said I wonder if there was something to how they cleaned the data because 65% performance on a feed forward ANN is phenomonal. usually LSTMs only perform like 54-56% at most with a bit of work (but the papers I've read use intraday data). I'm curious also why they published if there model performed so well. why not use it themselves if they get good results?

I dont know what the authors did but ill reiterate from before vanilla LSTMs do little better than guess on the stock market. They probably had like 1000 GPU and tested thousands of hyperparameter configurations and "overfit" the test set. This is why typically papers like this are not believed anymore in the ML literature. Try adding some stuff like attention or skip connections and whatever else is hot now (I'm not sure) and didnt someone recommend GRUs instead. I have some other ideas you can use like Gaussian Processes to estimate realtime covariance matrices, but your better off looking at the literature first than trying out hairbrained ideas that might not work. It's really not a trivial excerse to outperform the market with ML. Ah, I think I see where there is a miscommunication between us. :) My goal is not to outperform market with ML. My goal is to understand whether what is proposed in this paper works and if it does not then why. 8) You are possibly completely right that what authors propose does not work. I just want to understand how they got the results they got. You've said "probably had like 1000 GPU and tested thousands of hyperparameter configurations and "overfit" the test set." before. I don't think that's the case at all. If you read the paper - which you haven't so far - you can see that their training is really simple and there's no "thousands of hyperparameter configurations". Which is baffling in itself. I have some suspicions of what could be wrong, but it's not productive to discuss it if you just dismiss the paper offhand. Which is BTW your prerogative - if that's where you stand, that's fine and I won't bother you with this further. 8) You are entirely correct that I haven't read the paper and maybe I was too hasty in dismissing the paper. I wouldn't mind a copy of the paper if you don't mind sending me one. That being said here is my reasoning in more depth. The authors seem like they are in ML acadamia, so I made a couple assumptions. 1.) It didnt look like their paper made it to one of the premier conferences and maybe its because they aren't big names but likely its because people have been training LSTMs on stocks for a long time and vanilla LSTMs dont work well and I think everyone in the ML community is suspicious of 80% hit rates using a vanilla LSTM on indices for good reason and they likely didn't do anything special to assume that they didn't just get "lucky" with their model. the reason they got "lucky" is number 2) typically papers dont discuss the hyperparameter search they go through to find the exact correct configuration, so even if they didn't say they tested 100s/1000s of hyperparameters they might have and likely did (although yes i didnt read the paper). Unless they specifically say there were few or no hyperparameters to test or they tested only a few of them, you should assume they did test many. This is a dirty secret in ML, you come up with a new technique and you dont stop testing hyperparameter choices the model until you get good results on both the test set and validation set. Then you submit to to a journal saying this method did really well because it outperformed on both the validation set and test set. But you stopped right after you get a hyperparameter choice that met those criteria which strongly bias your results upward. This is related to p-hacking. This is a perfectly natural, but bad thing people do and usually means most papers have performance that can't be matched when trying to reproduce them. You can pick basically any method of the thousands that have been proposed and if it doesn't have over 1000 citations (and the method actually seems useful) this is probably one reason why. Now you maybe you are right and something else may be missing, but if I had to guess I think its a good chance the authors just got "lucky". BTW why dont ask the authors for their code. Its customary to either give this stuff out or post it on github. As a side note: Even a vanilla LSTM has many hyperparameters: number of states, activation type, number of variables to predict, test/train/validation breakdown, number of epochs, choice of k in k fold validation, size of batches, random seed, how they intialized weights (glorot, random nomal, variance scaling..) for each weight in the ANN, the use of pca or other methods to whiten data, momentum hyperparameter for hillclimbing, learning rate initialization, choice of optimizer... My point is that even with a vanilla LSTM the author can pull more levers than can be hope to be reproduced if you don't know absolutely everything maybe even down to the version of python installed to reproduce the pseudorandom number generator. No doubt some of these choices will be mentioned in the paper, but many of these choices won't be typically, which makes any reproduction difficult. And typically the authors are the only ones who are incetivized to keep trying hyperparam configurations until one works. The real papers that are sucessful are typically methods where either its not impossible to get a reproducible and externally valid hyperparamter configuration, or something that is relatively robust to hyperaprameter choices.

I dont know what the authors did but ill reiterate from before vanilla LSTMs do little better than guess on the stock market. They probably had like 1000 GPU and tested thousands of hyperparameter configurations and "overfit" the test set. This is why typically papers like this are not believed anymore in the ML literature. Try adding some stuff like attention or skip connections and whatever else is hot now (I'm not sure) and didnt someone recommend GRUs instead. I have some other ideas you can use like Gaussian Processes to estimate realtime covariance matrices, but your better off looking at the literature first than trying out hairbrained ideas that might not work. It's really not a trivial excerse to outperform the market with ML.

I dont think anyone is demonizing the russian people. It's the institutions and government that people are complaining about that allow those friends of ths government to become billionaires through no talent of their own. But that happens everywhere...no one complains about the same thing in China, India, U.S., U.K., Germany, etc. Alot of people around the world are billionaires with no talent of their own...some even become President of the most powerful nation on the planet! ;D Cheers! Sure but in russia most got there by knowing putin or yeltsin and many that lost there billionaire status did it because they pissed him off. This likely happens in china too (but even not to this extent) and certainly there is favortism in the US and other places, but you really haven't seen the world if you think the corruption in Russia or China (where I have lived although not in a 5 or 6 years) is equivalent to what we face in the US or Canada.

Here is there seeking alpha piece outlining their thesis (copied on outline.com to get around the paywall) https://outline.com/NkdYVn

I dont think anyone is demonizing the russian people. It's the institutions and government that people are complaining about that allow those friends of ths government to become billionaires through no talent of their own.

Yeah I have to some extent but don't read Russian. I would probably buy mail.ru in Russia because even though its sort of a corporate governance nightmare, it's such a good business and has some sort of stabilizing European influence compared to many of the other big Russian firms as well and so I think it will do well. But I can't by mail.ru so yeah.

I did quote the national review, but its easy to find newspapers where bush was saying he's liberating Iraqi's right before the iraq war and during. It's easy to find trump (like see a cnn article) saying the other countries are taking advantage of our miltary which is just a more cynical way of saying that aspect of our foriegn policy is doing the rest of the world a favor. Killing of journalist and oposition parties just ask Bill Browder also the recent attempted assasination in the UK. Putin dictatorship (duh I hope you dont need a source for that). Gregmail is right though the main issue with stocks though is the lack of institutions and corparate governence (again ask bill browder or follow amy company owned by Putin's govt or cronies). Do you think its a coincidencw that putin and all his friends become billionaires after he camr in power?

Not all of the businesses there are run by fascist dictators nor does Putin have a sphere of influence in every company. Furher, as bad as I think Putin is, has he really done anything that America hasn't? He annexed Crimea. The U.S. invaded the Middle East, supposedly, for weapons of mass destruction that were never found and destabilized an entire region in the name of "freedom" while killing more innocent civilians in their invasions and the ensuing wars than those evil dictators ever did. Further, we do unsanctioned air strikes in sovereign airspace, often times without the permission of the country we're operating in, which would be seen as a declaration of war if any other nation tried it. He meddled in the U.S. elections. The U.S. has meddled in dozens of elections globally and forcibly removes dictators it doesn't like, at will, while supporting other families/governments that are just as terrible but give us oil. Further, tapes revealed that the U.S. was meddling in the Ukranian election that resulted in the pro-NATO candidate winning which prompted Russia to act since "enemies" were now on it's border. I don't support Putin, but it would be hypocritical of me to say I won't invest in Russia b/c of Putin if I'm willing to invest in the U.S. despite it's "foreign policy". That's a false equivalence. We invaded Afiganistan because of 9/11. The US was attacked, they were harboring the terrorists. We attacked Iraq for similar rationale. We thought they had WMDs, we were worried that they would help perpetuate another 9/11 on the US. We were wrong on all those fronts, however we have withdrawn most of our troops and allowed self autonomy. While maybe Bush wanted the oil, you basically can't deny that to some extent he wanted to make the world a safer place. Conversely Russia attacked Crimea because they wanted to increase there sphere of influence and threaten a country moving closer to the West and expand territory. I concede that the US has done similar things in the past (Vietnam War mainly), but Russia is Machivillian to a degree the US is not. When we engaged in these wars we ultimately allowed some sort of self autonomy when we left whether we failed or succeeded (Korea) and didn't attempt to colonize these areas. The US does practice some realpolitik, but also has a reputation of supporting democracies like during the Arab Spring etc. Russia is neither a democracy and consistently (unlike the US which only sometimes) supports dictators. The US is neutral to supporting pro-democratic rebels in Syria, Putin supports Assad and his use of chemical weapons. Putin also basically controls the entire country. He assassinates people on foreign soil and at home who are a threat to him (even his own nationals). The US doesn't do that. I think the last assassination I have heard the US perform is Osama Bin Ladin and there is no way that is not justified. Perhaps you can call drone strikes assassinations, but these are typically done in war zones. Maybe the US assassinates people privately, but Putin certainly does more as we hear about public assassination attempts at least once per year. Additionally journalists and opposition party protesters have been jailed and killed for no legal reason. Furthermore, there is evidence that Putin bombed his own people to catalyze his rise to power. Sounds like a conspiracy theory and the truth is currently not known, but even a site with as much to lose (being affiliated with the Republican Party) as the National Review has lent credence to this theory: https://www.nationalreview.com/2017/02/russia-apartment-bombings-september-1999-vladimir-putin-fsb-cia/ Major newspapers run this as a fair and not out of right field analysis of how Putin came to power. The idea that the US and Russia are equally culpable is a false equivalence. Whether you agree with it or not, the US spends lots of money on the military and nuclear umbrella so that we can defend our allies from Europe to Japan. If you consider that being taking advantage of, you basically are conceding that we are helping our allies at our own expense. One big reason there has not been a global war since the 1990s is because of our enormous military which we use to discourage extremely aggressive and predatory action (even if sometimes it still happens and we still practice it). Basically yeah sometimes we are Machiavellian in trying to get what we want in foreign policy, but we also spend a lot of money and effort doing good in the world which is more than you can say for Russia.

I second this. Long stuff that is heavily dependant on tar sands economy but not E&P so no movement in responce to surging oil yet.

You could try a reinforcement learning approach rather than just a supervised learning approach. The upside here is the algorithm could learn to deal with risks and optimize a portfolio. The methods discussed in the openai post TRPO and PPO are very powerful both theoretically and practically and PPO is really easy to implement. I don't know reinforcement learning in depth. I wonder if there's enough data to run RL on stock prices. Unless you do it on intraday pricing, which I don't really want to do. I think it's the same issue as with supervised learning: 10 years of daily data is only 3500 data points or so. With only 2-3 crashes in data set. But I'd have to read up on RL to see if there's a way to apply it. If/when I have time. Thanks for bringing it up. 8) The best returns come from intraday data algorithms. Not the fundemental type analysis we are all used to. The reason is these algorithms may be able to average like 10 basis points after costs (just an example your algos probably arent that good). But if your holding periods are a couple of hours or even minutes, you can make 100%+ in a year which is just not attainable with any longer horizon algorithm.

You could try a reinforcement learning approach rather than just a supervised learning approach. The upside here is the algorithm could learn to deal with risks and optimize a portfolio. The methods discussed in the openai post TRPO and PPO are very powerful both theoretically and practically and PPO is really easy to implement.

Just cross validation during bull market years? I've played around with it a bit but never been comfortable enough with the algo [even worst with NNs]. I'm very scared of blowing up with these over-fitted models that have only seen rising markets... I think the main criticism against these "paper" strategies is you have 1000s of academics looking for signals and the winners publish a paper. The signals they find basically are the result of survivor bias. Do you guys have slack? Maybe its time we start a CoBF slack group So I spent a bunch of time reimplementing what these guys presumably implemented. I do not get their results. My results are pretty much at level of random guessing. It's quite possible I am not doing something the same way they did. As I said before, I'll link to their paper once it's publicly available and someone else might be able to replicate their results ... or not. I may also post or send my implementation to anyone interested after the paper is publicly available so people can shoot holes in what I did... Although I don't promise to clean up the code hugely... Right now it's a prototype-level mess. 8) The dirty secret in AI research is everyone is secretly overfitting their ANNs by by fiddling with the archtecture of the model and peaking at test set results. Only the papers with actual impressive results get published so you have a publication bias. Doesn't mean a lot of techniques don't work but they likely don't work as well as the paper would lead you to believe.

The shareholder base turns over as hedge funds and eventually individual investors realize they can use a loophole to wrest monetary control away from the government. But let's call a spade a spade and not delude ourselves into think we are crusading for the little guy and fighting for justice and maybe the spirit of the law (although we can disagree on that third point). I want to respond but already said I was done. gl! If you want to dm me we can talk about this in private (or we can start another thread on the politics board), but I don't really understand why you are bringing up my comments on other threads on this topic.

The shareholder base turns over as hedge funds and eventually individual investors realize they can use a loophole to wrest monetary control away from the government. But let's call a spade a spade and not delude ourselves into think we are crusading for the little guy and fighting for justice and maybe the spirit of the law (although we can disagree on that third point). I want to respond but already said I was done. gl!

Thanks. Good luck with this investment!

Look there is what is permitted by law and then there is what is ethical. Controversial statements: I think the legal points don't matter ethically and here is why. If the government was any other market participant, they could have gotten away with owning basically all the stock and all the bonds for much less than the 200+ billion dollar injection they put in. The reason the terms were so favorable to the equity holders is because Obama would have been called a socialist if he actually nationalized "the mortgage industry". So here you have the government granting more concessions than any other rational self interested market particpant would ever dream of giving, and then you have hedge funds who want to extract even more money from the government by expiditing the process in favor of stockholders and prevent the government from recouping a proper return on investment. This is why people hate the financial industry. The government saved your ass. Gave you more concessions than they had to and then when you recover you turn around and use those concessions to force the government to hand you a quick buck. Why do people hate Hank Greenberg for being an ass? Why do people complain about big bank bonuses? Why am I completely baffled by the self righteousness of this thread? Its because you may have all the legal justification in the world, but that doesn't mean it is the right thing to do and that is what infuriates people. You play "fair" but you don't play fair. I'm not arguing that markets worked this way and you weren't playing by the laws and regulations, but we also have a higher obligation to doing the right thing. You can't codify what is right up to every detail in a set of laws, so we have to rely on people doing what's right. I'm not going to complain if you think that Fannie Mae is a ethical enough to invest in, but when you start to act self righteous about basically extorting a government whose hands were tied politically, a government whose actions basically prevented a new great depression, that they owe you money after they saved the company that you or your predecessors were invested in, that kind of pisses me off. ok seriously flame war has to stop. I'll give you one more response but I won't reply.

Normally I would not say anything, especially since you seem to have no interest in the stock. However, this is a much deeper legal and accounting issue than two minutes of research would cover. Either way (i.e., NWS aside), a winning argument for plaintiffs is that the GSEs have been de facto nationalized without compensation of the owners (shareholders) in violation of the fifth amendment and centuries of international legal precedent. This complaint has been filed in the US Court of Claims and will be heard toward the end of this year. Let me summarize my opinion: you are wrong, wrong, wrong. The people who post to this board have invested considerable time and talent on the legal and financial analysis. I do not place myself in that class, but I have spent a fair amount of time on trying to understand the nuances of this investment. Finally, if one invests one's money in an asset, one deserves to profit fairly from taking the risk of the investment. That is the way that markets work. If the taxpayers don't take the risk of owning a security, then they do not deserve to profit from the ownership. The recent consensus (including even a recent Washington Post editorial) is that the Treasury Department has been paid back in full and even profited from its actions regarding the GSEs. By many assessments, the actual profit has been very large (100% or more) of the actual funds expended by Treasury. Even then, one could argue that the actual risk taken on by Treasury was nil. Not so with private shareholders, who invested their personal, finite assets and who cannot simply print more money. @rros for your latest comments. Maybe I'm trolling now too idk. But look I'm not even arguing the merits of the case. The loophole that Perry capital and others found may be rock solid legally, but here is what I see: Government saves fannie and freddie at an enormous expense and places the instituions into conservership (which credit default swaps consider to be a default). Stockholders play no role in the rescue as they wouldnt be able to issue enough shares (not even close) on the public market to stay afloat. The shareholder base turns over as hedge funds and eventually individual investors realize they can use a loophole to wrest monetary control away from the government. People on CoBF are indigant that the government doesn't give them the dividends they think they should be owed by Fannie amd Freddie. Look I'm not saying I'm not willing to go into ethical grey areas for a good investments as if I were alive and knew how well philip morris has done over the past 30 years I wouldn't have hesitated to pull the trigger. In fact I may have owned PM at one time. But let's call a spade a spade and not delude ourselves into think we are crusading for the little guy and fighting for justice and maybe the spirit of the law (although we can disagree on that third point).

I don't mean to sound snarky, but it took me 2 minutes to find this. The investment thesis might be sound but perhaps the confirmation bias has lead you to believe the right thing to do is what makes you money? https://www.frbatlanta.org/cenfis/publications/notesfromthevault/1405 I don't think you sound snarky, more like a troll. Nothing in that article supports your contentions. And yes, you are going to need more than 2 minutes of research. And most likely, develop your own analysis while stop reading other people's opinions. May lead to better judgment. Again I'm not trying to start something, but I find it funny that you think I'm a troll when a.) I post with my real name, b.) I have lots of posts both positive and negative about other stocks but c.) I disagree with you. IMO, I would look in the mirror and ask if you are guilty of confirmation bias (and writing off variant views as trolls). Personally, I don't invest in the stock because while I feel like preferred are likely to make money (again based on little research), but it leaves a bad taste in my month profiting off taxpayers knowing that public servants can be manipulated because they want to win votes and support. On to my arguments. Here is the paragraph in the article that I think supports my contentions: What this is saying is that before the third amendment, FHFA is arguing that the dividend was actually the result of waving the fee that FHFA is due (as they weren't profitable until 2012) and that the real agreement required full compensation from Fannie and Freddie. He then defends this entire position through the rest of the article. While I also linked the court case, you claim that the court did not rule on this profit sweep. From the title of the article on the court case, it sounds like the judges had an opinion on the profit sweep and it influenced their 2-1 ruling for the government. Even if what is true, clearly the judges could have ruled on the profit sweep, and if the ruling is a slam dunk in favor of investors, the judges probably would have ruled in favor on that point 3-0. At best, according to you, they deferred from ruling, suggesting that they think the argument "does not stand a chance" is likely flawed. I think your criticism is a little silly that I should do more research as I'm never going to own the stock, but I concede that maybe I shouldn't post on this thread if I didn't do research. Even so, I post here because I find the people saying that the government owes them the dividends a little self-righteous. Shareholders did nothing to keep Fannie and Freddie afloat. The government did all the work (also a point made in the piece I linked). The main stockholder base brought in after the fact because they thought they could use a legal loophole to force the government to give up control. It doesn't seem to me like current stockholders have any high ground to claim. Although I'm not disagreeing that you can't make money. Just because you can make money doesn't mean it should happen (that's all I'm saying).

Also says right here "Court reject hedge fund claim...in profit sweep." I could not be reading carefully or could be a bad title but sounds like the court had an opinion on sweep. https://www.housingwire.com/articles/39277-court-rejects-hedge-funds-claims-in-fannie-freddie-profit-sweep

I don't mean to sound snarky, but it took me 2 minutes to find this. The investment thesis might be sound but perhaps the confirmation bias has lead you to believe the right thing to do is what makes you money? https://www.frbatlanta.org/cenfis/publications/notesfromthevault/1405

I don't buy it. Both companies went into conservatership and the government spent 238 billion dollars backstopping both businesses. This doesn't include the more than 100 billion dollars that were spent buying debt and their mortgages. At the peak their combimed market cap came out to somewhere around 170 to 200 billion dollars and combined they were no where near that value during the recession. At the time, the New York Times said this bailout "could turn out to be the biggest and costiest bailout of private companies ever." Sure the government made out like bandits after the fact but thats like punishing WEB for quadrupling his money bailing out BAC. The reason the government didn't take a 100% share is not because they couldn't people would have sold their shares for pennies on the dollar, but because it reeks of socialism when government nationalizes a business and thats not politically feasible. Thus I think it was well understood that the government owned the thing. Maybe they shouldn't have rewritten the contract, I dont know about the thesis on the thing, but I think everyone was aware that this was basically a nationalization in everything but name. Your opinion or Allnatural's about the merit of the bailout -whether justified or not- are really irrelevant including anyone's opinion on whether shareholders should have been part of the recovery or not, never deserving a penny back. I think the central issue Allnatural is trying to raise is that the success of the bailout was based upon agreements. Since you like to base your arguments on law, rules and regulations or basic agreements (bankruptcy code), his/her central point -and the shareholders' fight in courts- is that those original agreements that were conducive to a successful outcome were not followed 100%. There is this one little wrinkle, the nws, that was not part of the original agreements that led a multitude of investors/shareholders buy equity in the companies. This minuscule change came about 4 years after the bailout when both companies were already posting profits making everyone who invested between 2008 and 2012 based upon the original agreements that sustained the companies as privately owned with residual equity to be had, a bagholder. It is really too bad that people like you are not happy that the original agreements and their authors could not foresee a happy outcome for shareholders (but many here did see it). We were on the road to become whole, all based upon government agreements when this little tweak attempted -successfully to this date- to rob the upside from us. 1.) The D.C. circut of Appeals basically rejected this argument 2-1 regarding the illegality of the third amendment, so i dont think you or allnatural are really telling the whole story 2.) From what I have read, the origional agreement said that the government had a right to change how they were paid "to fully conpensate the Purchaser...". Perhaps you can argue that government is compensated well enough, but I don't think making whole the government side is fully compensating given the massive risks the government had to take and the time value of money. Again just something I'm researching and maybe at some point people who bought this will get paid I don't know.