How Financial Firms Actually Make Money
Вставка
- Опубліковано 28 вер 2021
- Today we discuss the common misconceptions of retail traders regarding the algorithmic trading or more commonly termed ‘algo trading’. Here we dive into why finding alpha in today’s financial markets is non-trivial, and conflicts of interest present in financial markets.
Later in the video we discuss 3 trading strategies that are used as the foundation of most financial companies that participate in the financial markets. These strategies consistently make companies money and are the bedrock of companies providing financial services, or products in general.
Let’s aim to be pessimistic when people online promise that someone/anyone with a novice level of python experience can create an algo trading strategy that consistently makes money. I will endeavour to never give in to this fallacy and aim to point it out to my viewers on this channel.
★ ★ QuantPy GitHub ★ ★
Collection of resources used on QuantPy UA-cam channel. github.com/thequantpy
★ ★ Discord Community ★ ★
Join a small niche community of like-minded quants on discord. / discord
★ ★ Support our Patreon Community ★ ★
Get access to Jupyter Notebooks that can run in the browser without downloading python.
/ quantpy
★ ★ ThetaData API ★ ★
ThetaData's API provides both realtime and historical options data for end-of-day, and intraday trades and quotes. Use coupon 'QPY1' to receive 20% off on your first month.
www.thetadata.net/
★ ★ Online Quant Tutorials ★ ★
WEBSITE: quantpy.com.au
★ ★ Contact Us ★ ★
EMAIL: pythonforquants@gmail.com
Disclaimer: All ideas, opinions, recommendations and/or forecasts, expressed or implied in this content, are for informational and educational purposes only and should not be construed as financial product advice or an inducement or instruction to invest, trade, and/or speculate in the markets. Any action or refraining from action; investments, trades, and/or speculations made in light of the ideas, opinions, and/or forecasts, expressed or implied in this content, are committed at your own risk an consequence, financial or otherwise. As an affiliate of ThetaData, QuantPy Pty Ltd is compensated for any purchases made through the link provided in this description.
Every trading educator ever:
-"Every other Strategies out there do not work. Read my book/take my course instead"
TLDR; He doesn't talk about winning strategies. He talks about fees and bid-ask spread.
So is it not that impossible to make money algo trading as this guy has us believe.
@@omairtech6711 its 100% possible. Banks dont trade manually. they trade our money with bots.
That’s why he gave you the warning why UA-camr Traders are fakes.
Bad info, bad analogy. Maybe learn more about both before you make a video giving bad info on stuff like this. It almost amounts to a red herring.
His secret Alpha was to entice us to see the video from start to finish to avail UA-cam monitization.
This reminds me of the old story about how selling shovels and mining tools to the gold miners in the 1800s offered more guaranteed success than to actually mine for gold. I suppose the same could be said for many industries: Publishing companies earn more money than authors, youtube as a media platform probably makes more money and has a more sure thing business model than an individual who starts a youtube channel.
brokerage firms earn more money than most successful traders
So your advise is for all of us to either open brokerage firms or quit trading and starve the brokerage firms into insolvency
@@nwoksh It's an insight, not a call to action
UA-cam has always lost money actually.
How The Bible led me to Islam ua-cam.com/video/aFMHH9Hp-OM/v-deo.html
Love your video! Much appreciated.
I also love that book.
BT overfitting and ML.
Backtesting is like cash
The teachings on this channel are always top-notch, informative, and easy to understand. It's very hard to find good content online these days
I agree with you on everything, these days finding a financial mentor is a tough challenge which is why I am happy and grateful to have been introduced to my mentor Larry Kent Burton by a friend i made a lot of money in just two months working with him for just a small investment
Who exactly is this Mr. Larry? what does he do?And how can I take advantage of him
Sorry for interrupting your conversation. I just wanted to add to everything, Mr. Robert is a miracle trader. He helped me grow my trading account from just $ 1000 to over $ 11000, I can confidently say anyone who invests with him, you are guaranteed to make profits
He is a financial advisor and investor. He helps people to better understand the financial markets, and he also does trading and investing on his or her behalf
Getting in touch with him is very simple. Just follow him on Instagram
Thank you for the great video. This is one of the clearest summary about the business models of the large institutions. I’m subscribing because I want to look at your older videos on python and ML. But I agree with Sarahvandistel though, the difference in target niche between institutional and retail traders means different decision-making strategies, hence computational models, should be used. I still think that algo trading for retail traders is a viable method. We just need to find the right strategy. Once we get rich we remain quiet about how we get rich 😂, like those who already figured it out
Your videos are like a rocket ship to success! Thank you for guiding us on this journey.😄
Love your work. This is the first video that prospective traders should be watching.
good stuff, keep it up!
looking forward to more of this high-quality content
Man I love this video man! you are truly good at explaining what a lot of people think is complex.
Really good and instructive video on financial services as a whole. That being said, I think the algorithmic trading tutorials you're referencing are still useful for getting from 0 to 1. That is, how can one get to a point where you're actually analyzing data and coding primitive trading strategies. Once you get from 0 to 1, then you have the tools you need to research deeper and more complex machine learning strategies.
Congratulations for your work! Could you do a video diving on market making? I put a series of questions under your video "understanding market makers" that you could use to make another one diving into this subject.
How can a market maker (MM) make money with the bid-ask spread when you have many market makers acting in the market?
How can the information of the order book be used to make money for MM?
Why should they buy order books from other brokerage firms if they have already access to the order book of the stock exchange?
This is still unfortunately a big black box! Thanks
Yes, I’ll make a video on all things market markers at some point
An interesting rabbit hole is the channels financial incentive is UA-cam Ad revenue. If a video including "algo trading is fake" gets more views than an algo trading strategy video, you have to wonder are the strategy videos fake, or does calling them fake get you more views therefore more ad revenue?
Thats a pretty good fucking question. After watching half-way through this video I realised that I shouldn't be listening to people talks so much. I should spend more time working on my trading. Only experience and time will tell you what is the reality of trading algos. I will continue to develop my algo no matter what shit head on youtube says and see what happens for myself. I'm taking this video with a single grain of salt.
@@buyhighselllow8398 Seems like you're taking it with a whole mouthful of salt lol. Why so mad?
@@RyeBreadSucks I'm not mad, just frustrated that I haven't progressed with my trading because I'm always in the search for answers on UA-cam and end up wasting my time watching videos like these that discourage me to do my own due diligence. The main message I got from this video is that algo trading is extremely hard and that there is almost zero chance of making a decent algo that gives you consistent returns unless you're an institution with big pockets.
I hear you it takes alot of time trying to figure out and decide what to do and what to believe is a true way to make money and not a scam kind of scheme to get money from you
Love this video, I was searching for SOR and algorithmic trading on internet, thanks for sharing reality.
Thank you for the video. It's always good to remember that the financial market does not generate the money, it redistributes it. And if you don't see the one who is going to lose their money in the deal, it's probably you.
Thanks for keeping it real bro and not trying to sell me anything!
you are the product being sold
Have you got any forums for beginners to ask the trading coding questions?
You are recommending a book written by Marcos Lopez De Prado.
Marcos write very advanced books for audience with PhD.
Most of the traders don't even know how to spell their name.
Thank you for the thoughtful content! Very refreshing. I was beginning to think nobody but gurus and quacks created videos in this niche
brother I need your corner desk!!! Where did you get it?
firstly thank you very much for your tutorials. Can you indicate a book or training focused in technical analisys? I'm a beginner and I need to learn more about it to be a good operator on this market. Thank you!
I am a huge fan of your channel, Mr. QuantPy. Concise explanations and a great accent!
Fascinating video...
Thank you for sharing...
Always good to share knowledge and bring light to a topic...
I would like to see maybe how these python ML products work? Just a simple demonstration. Can you find the ‘alpha’ with python? It is a curious question - can statistic modelling produce the desired outcome? Carlo
New Subscriber - great video - a fellow fan of Dr, Prado
those were not profitable trading strategies those were profitable business models..misleading title
Thank you for being the voice of reason. I only wandered down this rabbit hole because I work in the sector and the implications of algorithm and api programming calls is an interesting subject.
API programming calls? First time I hear someone talk like that lol
@@dima6488what surprises you about him saying that, I guess it is not technically correct?
@@dataanalystbynight4375 Because no one who actually works "in the sector" (i.e developer) talks like this, since "Application Programming Interface Programming" doesn't make the slightest of sense
How The Bible led me to Islam ua-cam.com/video/aFMHH9Hp-OM/v-deo.html
Curious what you think of guys talking about support/resistance, breakouts, scalping, etc. A lot of it seems harder to condense down to a fixed set of rules (and therefore difficult, if not impossible, to backtest), but you hear a lot of the same ideas coming from different people.
I am currently living off of my Algo trading gains. Took close to a year and around $27,000 in losses before I figured out how to become profitable. It's not as easy as it seems, but it's not entirely impossible
Thank you you're giving me hope I'm currently looking forward to learn python probabilities and finance to get an algorithm working. Can I ask about your studies or how much time did it took to be successful ?
@@C-ly-demathematical when person claims he is profitable in reality means he is not.
@@paulnyagini yeah ig sooo...
your work is highly informative. I always do ask myself though - are you speaking as though only quants make money in markets consistently? I would argue alongside math phd quants you have non-quant traders in existence - seems as though you are not acknowledging those
Great video. You should make a video or series about the history of alpha. What sort of alpha strategies have been used in the last 10 years compared to the last 50 years or something.
Great suggestion, I will try and work this in soon!
@@QuantPy thanks you for your great work 🙏
@@QuantPywhere is the video? Have you made it?
One of my collegue who was also interested in joining some platform has asked for guidance for following issue
Dear Sir /Madan
Good Day
In order to improve the efficiency in execution of trades and to avoid the slippage i am considering to work with
some Algo trader platform .
We have to provide the API from broker for integration into the Algo system .
My worry is that since the algo trader will have access to my account for placing the trade , he may
sell deep OTM options to me and my capital will be eroded instantly .
Please guide me if there is a possibility of happening such instances and what action and precautions i should take .
Awaiting your valuable suggestions for improving my learning curve .
With best Regards
Simran seth
I have tried to make a stock market sim but because YT is flooded with algo trading tutorials (which I don't even want) I can't find the content I want to use.
Good vid. Nice to see another Aussie. Subscribed.
Your information is very comprehensive, and I have learned a lot. I am in the retail trading business, but I know very little about investment banking. Thank you for sharing, I'm starting to understand a bit more.
Excellent video, and thank you so much for your great channel.
Title is misleading. The 3 strategies proposed to make money have nothing to do with alpha. But OK to know.
Your trading strategies have helped me make consistent profits. Thank you for sharing them with us!
Interesting video. I've been trading 3 years, and I find it interesting that you and others think it's impossible to find reliable alpha. Nothing is absolute, but direction is still a 50/50 break, which is better odds than many places. Plus, there are many different areas in the market, all of which follow patterns. However, what is scary is the ability of market makers to swill in the info of retail buyers before essentially going against the majority of them. That does seem to possibly be a bridge too far.
Is there a way to arbitrage the market? I am very new, but I know with sports betting between two bookkeepers you can create arbitrage bets against each of them. Can the same be done in trading? The context is obviously very different.
Yes but not as a retail trader.
You'll need to be a price maker/taker across multiple institutional venues and exchanges, operating under institutional execution terms.
Most retail accounts for example won't let you bid or offer inside the prevailing spread.
And most retailers won't be setup to buy in Venue A and sell in Venue B. Institutional price aggregators offering Prime Execution Services open up such opportunities to professional market participants.
What most retails traders think that Alpha is just the excess return over the benchmark, whereas you have to consider risk adjusted return to compare it to benchmark.
It is not alpha, if you beat the benchmark with over exposed and leveraging into riskier positions. And yes, it is not consistent as well.
Very much agreed! So many algo-Tubers never show the MAE of their trades, nor show e.g. Sharpe Ratios.
Estimating Alpha is an ML problem.
An additional benefit to “quant” trading that has nothing to do with Alpha is the elimination of emotion: as long as you can truly follow your systems. Having a system can be as simple as never going long when the asset you are looking at is below it’s 200 day moving average. This sounds almost idiotic in it’s simplicity, but I would say that 90% of all people I have shared this technique with have been unable to follow it, let alone anything else that is more complex. Following a simple 200 day rule could be the first step in reducing volatility and drawdown in your portfolio, thus making it more emotionally easy to actually stay invested. I’ve used this technique (along with a lot of more complex ones) as part of my toolbox for over 30 years and it has helped me out significantly.
Significantly in trading = ££££. 30 years , wow. You should be in millions , at least.Or managing tenth of millions.Yes , the 200 is a known thing and actually 300 seems to be more accurate. But there is also let's say market structure, that can offer you opportunities. Trading is a complex thing and people are different so there's no universal solutin .As about algo trading , I think a statistic several years ago showed how ''human'' traders had better results. Also saw a docu about an algo firm with some very bright guys that somehow went bust .
That's funny because I'd say buying below the 200 day moving average means that you're buying at a fire sale.
Traded for couple of decades - 10K transactions/yr; Liquid markets are mostly random/efficient, so you can buy or sell above or below any moving average... It's still a 50/50 bet; You may win, but it's not a viable business model. Best wishes in trading and investing!
@@cooltrades7469 A couple different studies I've seen found that most technical indicators slightly improve performance when using longer period inputs.
Market regime changes are what mess up most traders when they try to apply a single tool or toolkit to all situations.
Re: algo firms going bust, there are plenty of them in the past two decades b/c most of them create and test a methodology that works at inception, but that market edge is perishable and erodes over time if they cannot sufficiently adapt and update their methodology to follow market conditions over longer timeframes. It's the same tool limitation problem, writ large with hundreds of millions of dollars on the line.
@@bubbleboy821 Depends on whether the slope and concavity of the MA are positive or negative.
If your MA is sloping upward, and especially if concavity is positive (showing upward acceleration), then yes, buying the dip is great.
Downward slope and especially negative concavity would mean buying below the MA into an accelerating downtrend -- better to sell. John Maynard Keynes himself said the markets can remain irrational longer than you can remain solvent. :P (I take issue with that quote though, since the main part of the fault lies with us for irrationally pitting our ego against the market's direction in the first place)
so few cuts in the video! respect ♥
I started backtesting some strategies with years of data on a 5m ohlcv timeframe. Got some strategies which make me around 800% ROI each year. And like each month almost the same amount of winning percentage.
Still developing as a side project a real stable live trading bot to take these strategies into the practical world but I am very curious what you think about it? Would there be a high chance of this being over fitted for historical data or could it actually work and make profit because I backtested it with lots of data?
SO you can make 8X every year? I will wait until your Net worth be a Million dollar & you open A Hedge Fund Firm.
Iff you can do that I will give you 1 grand for your Bullshit 😂😂😂
Hey I'm happy to hear that for you I want to get into algorithm trading and wanted to know what study did you do and how long did it took you to get where you are
800% ROI with consistent win rates each month? what's your sharpe ratio? 8? lol of course it is massively overfit.
Good work. Pretty much agree with you. Finding alpha is hard but not impossible. Technically speaking retail can do market making in the crypto.
How do you get initial liquidity for the AMM and compete against other big players such as uniswap?
@Xunning Liu I was talking about market making in centralized exchanges like binance. Making AMMs are not retail friendly.
If you’re paying ‘gas’, you’re not market making😂
Im sorry but referring to econometrics as 18th century tech just because (some) techniques involve inverting a 'simple' covariance matrix is a bit of a stretch. Most of what people label as machine learning nowadays is actually based on 18th century math, eg. PCA. The author of that book is openly against the use of econometric techniques (against ML) which is only one side of the story... peace.
That red book is the intelligent investor, what are the other books on your shelf bro. And I think I've seen a pattern that can make me money, I'm in the process of coding it from scratch. It's taking a while, a lot of these things are new to me, but I've finished with market data collection, now for pattern analysis and decision making
Ah the bid ask spread can be an issue, I've learned something thanks.
Thanks a lot for this valuable video. What's your idea about algorithmic trading in Cryptocurrencies? They actually don't have intrinsic value and some economic terms are not applicable.
Hello man, I have some question, Is quant profitable for retail traders?
For less than 1% of retail traders - yes.
The issue is 99% do not have the patience, skill, discipline or knowledge to research and develop good quant strategies.
Most professional quants themselves are not profitable and many funds on balance underperform a naked bet on the S&P 500.
To get rich trading you need long term consistency and a massive fee-earning bank roll. Without a bankroll the vast majority remain over exposed (over leveraged and over trading) leading to guaranteed account blowout.
Long term consistency is very difficult if you're doing it manually - it's orders of magnitude tougher programming a computer to do it.
These are the simple facts.
The biggest issue with most quants is lack of creativity.
You can't learn algorithmic trading from a book or course.
You research and develop these things through methodical experimentation and trial/error.
Even then, most strategies do not stand still - they're revised and evolved to address constantly changing market regimes and sentiments.
You absolutely can find alpha as retail. Especially true for liquidity constrained strategies. Large funds are not going to bother making 200-500k annually from a strategy, it's way too small for them. For single trader retail it's worth it.
you are 100% correct I am a small retail trader and I only trade alpha, this guy doesn't understand direct access brokerages as well it seems, I have only watched a few minutes but its enough, changing the channel after this post
So, in other words, if you go into business for yourself and learn the business, to use the analogy from the book this person references, you can metaphorically mine a small portion of gold dust but, for a one person shop, that profit is worth it? (That is a genuine question, not a back-handed rebuttal)
@@jerryc5743 Exactly. Not everyone in the marketplace is in direct competition with everyone else.
@@incremental_failure thanks 😀
Are you speaking from personal experience?
hi is there any room to develop winning strategies for trading using GPT by developing AI bots?
Thank you for this video. There so many brokers out there - only selling crap! I’ve seen so many trading academies selling some of special platform (only theres) are able to make money. There business model are just selling a "crap education packages" starting at around $2k to $15k. Carlo
So is intellectual knowledge in bit coin or cryptos Alpha
Got it - don't bother chasing alpha. Is there a way as a retail investor that we can reduce beta with minimal impact to our 0 alpha? Or are there no good, easily accessible beta-reduction strategies either?
To be fair, following the S&P 500 or buying ETFs are the best method at reducing beta. They're simply the easiest and less risky methods, but of course this will reduce your potential income. There is no such a thing as low-risk high reward investment, especially not in the short term. Long term investments are an another good example of reducing average volatility. But of course if you're not careful long term investments can be even riskier. Most big corporation's price will go up eventually... but there are a lot that may never go up, especially if they the corp uses the wrong strategy.
19:44 he says it’s not impossible to find alpha.
as jack boogle phrased it: the only free lunch in the stock market is diversification
Dont be a beta, be a omega chad
Just wondering, if an algo that is backtested that seems to work, is there a good chance it'll make money live ? Anybody have any experience with that ?
Only one way to find out I guess, try putting small amounts at first
(I never did algorithm trading but I know sometimes you need to take a leap forward)
If your algo works (10 years or more) backtesing , you need to try it first on demo accounts live. If win in demo account then try with small amount real account. If it wins again its up to you :) I dont think algo will not work , when writing it you need to calculate all risks not profit. And it need be SIMPLE. Not complex , to work in different conditions. Algos in videos definetely not work even with backtesting ones :D Not copy paste them. But big benefit of algo is you can test your startegy.
Wow very good video, POV I didn't think of. So far from trading, I see that there a strategies that can help get that edge, but the best strategies are the ones that keep you in the trade longer while managing low risk. So for me (and this can change after writing this) that would be Supply and Demand, why? because that is literally what moves markets! Thank you for the video
So you saying is that brokers take retail traders data and provide that information to the market makers.
I had some time and tested out several HUNDRED financial theories, and not one of them worked as advertised, which should be expected. Why would anyone release an effective algorithm? Ironically my only working bot was the first project I worked on when I knew nothing. I saw a potential pattern while watching some chargers, and it turned out to be surprisingly effective.
So basically your bot is an algo that works.
can you share with us your pattern
@@chillbackground4935😂😂 I’m sure everyone has the same thought as you… and I’m sure he won’t. Reason one, he spent all the time and money to test the algos, he deserves to be rich. Reason two, once it’s public it won’t work any more. But he already gave you plenty of information, right? From his brief description, you can reduce your search for a working algo by almost a HUNDRED fold. All we need to do is to figure out his hidden message and do one-tenth of his work.
Hey I won't obviously ask you for the algorithm but I'm looking to use python for trading and wanted to know how much time and/or what study did you do to get where you are.
Have a great day
@@dawei6697 what I will say is that you have to check out the failures to see why they fail. And always always always use a benchmark. It’s near impossible to develop a working boy without a benchmark.
your strategies do work but it would be very nice of you if you make videos on how you study the candles and the market volatility/movement. Love from India.
The gold analogy is very good. And yes, nowdays alphas can't be found with a seive, they need to be studied with academic high-level data science approach. Just the same way these new gold mining machinery is designed and developed
The video is hard truth I guess for retail traders. But then in your view, do Retail traders have no hope? I have seen successful retail traders making good wealth? how to reconcile it? Thanks
Appreciate the book recommendation. Thanks mate!
Is there a big advantage of knowing mathematical finance for an individual trader? Also, is mathematical finance only for derivatives trading?
I would say no. I know plenty financial math from college and it doesnt help much for trading.
@@JuanDeDios24 thanks
@@seadotocancerdoto1565 the education that they give us is just for working for them.Nothing more or less. So you have to educate yourself by practicing.That knowledge is priceless and can't be found in any book.
I also think that there are other macro opportunities that present themselves when hedge funds turn a blind eye to value based on societal pressures. Presents an opportunity to individual investors who don’t have the same pressures (ie. Job loss/no promotion) for investing in non-green companies. Energy & tobacco for example.
Johnatan how is it possible to transfer cryptocurrency to your account for HFT traiding under your control?
Thank you for the thoughtful content!
Why strategy #1 and #3 are not the same? They are the exact same thing for me: fee for a service.
I got tangentially interested in algo trading through a random video suggestion a few weeks ago. I watched a bunch of videos and I came to the realisation that the only guarantee is that the house always wins, and this video confirms that. Sure you *can* make money on the markets, but it's not a get rich quick scheme unless you're a) very lucky (and then it seems people try to ride the luck and lose it all), or b) a very talented risk-aware person.
You didn't really talk about algorithmic trading but rather on the business modell of the two different groups (directional vs non-directional). Thanks
He has to seek the alpha of the UA-cam algorithm somehow haha.
Agree. Very misleading title. One doesn't have to know python. There is software out there that allows you to create your own algo and test it, which is far better than manual trading.
This is one of the best videos out there on financial education! I am going to make it a requirement for my students.
This video was 20 minutes of self evident stuff that everybody should know already: a) making profit is hard b) brokers and institutions make money from other things like commissions. If someone didn't know all this before watching this video, they should not be trading anyway.
Great video, kudos! I totally agree with you. If you are trying to use algorithmic trading tools and techniques to get alpha, you are probably wasting your time (and money). These tools and techniques seem to be much more useful for 1) monitoring the market (think TradingView alerts, e.g.) 2) managing risky situations (think stop losses) and 3) automating certain processes (such as re-balancing portfolios).
What's the basis for your opinion?
@@incremental_failureThis video (he says he agrees with the video). 😉
well, actually I use my own model and input manually. I will not write the code in any platform since I am afraid can be however in many ways be stolen
Good analysis-Bring it back to reality a product! Well explained- know amount of resources on earth 🌎 for profit-Alpha is my new knowledge thanks
I trade a manual algorithm the NNFX way and that works for me
I love your videos. It's something the internet needs.
So, guaranted IS not reachable by retails.
Guaranted IS also u tube, as far as WE Can use infos. Tks.
So what does the H stand for?
Very informative. Really tired of all the fakes. Keep up!
What do you mean that you can't create a strategy that spits out a sharpe ratio of 10 in 3 lines of code? But...what about my moving average crossover idea???
Does the Black-Scholes equation ever come up in quantitative finance? I don't know too much about the equation but its supposed to be the scariest equation in modeling time evolution of markets/risk.
Definitely, this has become the most well known simplistic model used for European option pricing.
@@QuantPy "simplistic" lol
@@albertociampini6473 I can't tell if you're agreeing or disagreeing. :D "simplistic" is a pretty nice euphemism for a formula that assumes and requires a constant value that can't be known in advance.
@Starvling k TLDR Blaming Black-Scholes as being "scary" is a scapegoat for human primal stupidity.
I liken Black-Scholes to learning high school physics -- you're taught to calculate examples like the acceleration of a frictionless, spherical chicken with a uniformly distributed mass of 2 kg, sliding down a ramp in a vacuum. You're taught this grossly simplified abstraction b/c you don't yet have the capacity to factor real-life variables into the equation: air resistance, flapping wings, shedding feathers, chicken pooping, running around headless, etc.
If you use that level of physics to build a rocket that subsequently blows up during launch, then out of ignorance blamed physics for being "scary", it'd be equivalent. The limitations of both Black-Scholes and that simplistic physics model were known and criticized by experts for decades prior, but you as the aspiring rocket scientist and hedge fund manager were too busy to listen. And then, what a terrible, tragic surprise.
I think the news media latched onto Black-Scholes as "scary" because some academic sound bites pointed to it as the cause of the 1987 crash and the 2008 GFC. Or it may be a case of upset people looking for a causal factor to blame after the fact of the crisis event. TLDR they blame the use of Black-Scholes for contributing to systemic financial industry shortsightedness in not detecting and robustly protecting/hedging against outlier long tail risk.
This is partly true -- Black-Scholes doesn't account well for long tail statistical outliers. Black-Scholes assumes constant volatility known in advance, which is not realistic. The volatility smile of options pricing shows that actual options pricing fudges the process with variable volatility -- implying high volatility for farther OTM options. The "smile" can also be skewed depending on situational bias to the upside or downside. A long tail event (market crash, whatever) is that one-in-a-million time when you pull that extremely high volatility lottery number.
More crudely and regardless of whether the Black-Scholes equation was directly responsible, I think the majority of finance types are possessed by basic human risk-aversion and the impacts of risk/uncertainty on their paychecks, to the point of blindness. Most people pursue the holy grail of eliminating or fully knowing all risk. If they can quantify it, they can price it into their revenue models. But they get sloppy and conflate 98%, 99.9%, etc known risk with "effectively 100% known risk". And that works for a few years at a time, until a 0.01% outlier event number gets rolled on the dice of fate, and then everyone panics b/c their revenue models failed to account for it. Increasingly interconnected electronic global markets since the 90s have made it more likely for local outliers to be felt on a global level b/c someone somewhere decided at some point it was a good idea to chuck some investment funds into a high-risk developing market to chase lesser known, marginally higher returns.
Thus during those years where the outlier event hasn't happened YET, week by week and quarter by quarter, every market participant is chasing down tiny percentage gains over their competitors, wherever an advantage can be found. That's what led banks to take on increasingly worse credit risks, then packaging those hot potato toxic assets up and shoveling them out the door for resale before the music stopped, leaving them holding the hot potato. "Everyone else is doing it profitably too, and the house of cards seems to be holding up just fine so far..." Because everyone thought their models effectively accounted for ALL risk, they pushed and abused the limits of the system, repealed various market safety mechanisms that were originally established after the Great Depression, and ultimately volunteered to become an industry case study in the Peltzmann effect (people are demonstrated to take on greater risks when they perceive they are protected adequately by safety equipment like seatbelts. Tragedies ensue when their PERCEPTION of safety overestimates the ACTUAL effectiveness of the safety equipment.).
@@dakaodo If the equation wasn't scary then everyone will be solving it. The only reliable solutions are stochastic (or monte carloish) based at best; those methods then end up proprietary.
Andrea Unger 4 time world trading champion. Kevin Davey also world trading champion, Bob Booker, Jarrod Goodman are all full algorithmic traders. Any comment on how that lines up with Alpha!
Exactly... this guy doesn't know how far off the mark he is. There is huge alpha to be found via algo trading (for the small scale retail trader). I personally have found it... and it is vast.
The big question is, why would anyone share alpha? If alpha is shared , the participants would use it and it probably would not be alpha anymore.
People are dumb. They don’t ask big questions. These programs are to make herds of dumb investors for the financial companies to use to move markets.
Would it become beta then?
@@don_huzan It becomes Pi. Pi for the financial companies to eat.
@@tvviewer4500 haha good one
This is exactly true…..I find it funny that people want to open trading academies to “teach everyone” thinking their edge will still maintain. The painful truth is not everyone CAN be successful….there needs to be losers for there to be winners.
You can’t tell that hedge funds are juste interested by managements fees
If they manage people money that because their performances and alpha are positive, that what bring more customers to trust them.
So anyway, they have to trade with people money and have a great trading strategies to make return.
What you can say, that those companies used others technics rather then juste predicting stock price. Some example are : hedging, statical arbitrages, means reversions, factor based models (using fundamentals).. and all of this drived by risk management
So yes even big companies have strategies and trade on market and make money with that
Well algo trading and alpha discovery are two different things in my opinion. The first can be as simple as, buy at price x and sell at price y. The latter indeed is not as trivial and has high barriers to entry. Good video, really enjoyed it!
no...
True, Interactive Brokers will let you as a retail trader use about six different ‘algo’ methods to buy/sell, but they really have nothing to do with alpha, only with different ways to enter/exit the market, e.g. not showing your full order size (‘iceberg’), attempting to buy below VWAP, etc.
@@ScottHz How would IB limit the kind of algos a trader uses? You are confusing some entry methods IB calls algo and not understanding what the word means in the context of typical algo trading.
@@cyberfunk3793who said anything about 'limit'? By 'let you', I mean IB provides them as part of their trading platform. That's why I put 'algo' in quotes, I know they're not the same ;)
If you look at the most successful hedge fund manager, Jim Simon, who has outperformed the market with an annual return of 66,1% before fees for his medallion fund from 1988 - 2018, it clearly shows that it requires the brightest minds in the world of mathematics, computer science and physics. The point is that most retail traders don’t have the necessary skills to analyze the right data and the right amount of data in order to produce strategies that can make alpha. Of course, that’s just how I analyze the situation. But most retail traders should definitely not use basic algo strategies like SMA 50 crossing the SMA 100
I get what u are saying but also some tools are as they stated and are just a means of you automating a strategy you have or future ideas. It’s not like they are giving you a strategy and saying it will work just because you automate it. It is YOUR strategy so whether it works or not is up to you. Those institutions just have strategies that work for them and they’ve automated it possible using their own in house tech guy and softwares. I won’t say a retail software will work as fast because there are a lot of factor but depending on your strategy just as fast may not be what you need.
I don't understand how can you completely ignore the buy-side.
I especially like trainers who talk about their crazy earnings and... encourage you to pay for their courses. It looks strange: usually those people who have money are not particularly interested in attracting all sorts of beggars and teaching them something.
Don't you like the UA-cam crypto or investments scam bot thread?
Oh the markets are really hard and growth but I'm really glad I'm still able to make $xxxx a week.
Then some one asks how and they name drop some unicorn guru with an email id or Whatsapp number in a funny format.
"Let’s aim to be pessimistic when people online promise that someone/anyone with a novice level of python experience can create an algo trading strategy that consistently makes money. I will endeavour to never give in to this fallacy and aim to point it out to my viewers on this channel. "
As someone who wrote their own python trading script I can agree with that, while it doesn't lose money it often misses good trades because conditions were not met, I certainly wouldn't pay for an algo bot even if I didn't know how to code.
He literally talks about how UA-cam videos about trading strategies are bullshit and then in the next breath offers trading strategies, without even missing a beat! 😂
Is hft also profit on bid ask spread, are they market makers?
HFT strategies are mostly front running, market-making or abitraging.
You can't successfully do it without access to institutional market data, futures exchange data, institutional client orders or retail order flows.
And you can't do it effectively without a massive bank roll to get started.
And that's before you set funds aside to pay for data access, hardware, engineering of the systems as well as the six digit salaries for the guys and girls that make it all work.
Just so we're clear - some HFT shops are doing well over 250,000 orders per minute across 1000s of securities.
Most indicators are trend following and fail in flat markets. They also not work consistently throughout all assets. I am searching for price patterns with a high win rate.
So you should identify the trend: long/short/flat and use the corresponding bots. For example, for a flat trend you can use a grid bot between resistance and support. For trending bots there are plenty of strategies using moving averages, lucid SAR, DCA etc. Once you identify the trend it's easy to make steady money just don't be scared to use stop losses: The trend is your friend until the end of the trend 😉
Goodness this is a refreshing island of sanity in the ocean of madness that is financial youtube
Advances in Financial Machine Learning seems like a great book thanks,.
hi there is also scientific publication which also describe moving average, markov chain etc.
Indeed, selling onwards live retail-position information would be lucrative, "easy money" , however why is that not "insider trading" ?
a bloke with his head screwed on! and he's an aussie too! awesome stuff mate
If you find a high beta stock (3>) and buy in a bull market or sell in a bear market (according to federal spending), surely you will find a significant annualised alpha since you are essentially playing 3x whatever the according market is.
- “Just time the market bro”
It's easier to show how hard finding alpha is by looking at hedge funds and which ones actually make money in the end. They are primarily after alpha.
And they are all Quant based on massive super computers.
@@HermanWillems not all, there are many 'stupid' hedge funds too ;)
Made an entire set of algo scripts, backtested them and verified their results, fixed all issues that are hidden in the field and finally got a morning making machine so that a guy on youtube can tell me its all fake?
Discord??
Does it work for you to make profits in live trading?
Same!... I have good profits in live too.