Australian (ASX) Stock Market Forum

Tools being used by your trading competition

Jump to new
tech/a said:
Skyquake

Your logic runs in the face of all Fundamental analysis as well.
Click to expand...
How so?

Quantitative judgement is not easy, if possible at all, to backtest. Despite the lack of 'objectivity' I will take an understanding of competative advantage (for example) as an edge any day over the best looking back test equity curve from the most sophisticated computing power available.

Ps if you actually read some of the BH letters to share holders you might learn something rather than just misrepresenting Buffett and what he has taught.
 
tech/a said:
I don't think this is an issue.
Retail Traders are using vastly smaller capital bases.
As such they can be very nimble with their entire account.

With the power and knowledge available to retail traders NOW and not before why wouldn't you incorporate it.
Click to expand...

Point is any hedge fund can go wide and trade in $5k parcels across 50,000 stocks. They can be nimble too.

Skyquake

Your logic runs in the face of all Fundamental analysis as well.
If retail can spot an under valued stock so too can the bigger players and the were doing it before computers as was everyone else.
Click to expand...
Buffet sized players will probably need to look at $1bil+ MC before they even begin. They can arguably use ML/algos to look at 10 x $100m co's or 100x $10m co's but still they'll need to review each one on a case by case scenario.
Certainly they can't have a robot talking to management and asking insightful questions
 
craft said:
How so?

Quantitative judgement is not easy, if possible at all, to backtest. Despite the lack of 'objectivity' I will take an understanding of competative advantage (for example) as an edge any day over the best looking back test equity curve from the most sophisticated computing power available.
Click to expand...
Just to clarify; do you mean qualitative? (see bold)
 
skyQuake said:
Point is any hedge fund can go wide and trade in $5k parcels across 50,000 stocks. They can be nimble too.
Click to expand...

Yes I get that-----you see bots plugging away all the time.
But 2 things come to the Ducks Potato head.

(1) With such a broad base aren't results less likely to out perform and more likely to conform with the returns seen in an/the index?

This is one of the reasons we don't see large funds out performing--there are a few yes but clearly they aren't doing what the others are.

(2) In the case of a sudden move they still have the same problem getting out in a hurry.

This can be seen when funds take large hits like everyone else. Some of the smaller retail guys get out before and I know a few here have when they have seen/suspected it coming.
 
Been a few comments made which leave me wondering if this is a Man vs Machine discussion.

In my view it is a Man vs Man+Machine (Learning) discussion at this point, but it isn't my thread. Ultimately it is an all-comers cage fight which requires that some of them stay fighting in order to actually have a secondary capital market.

Somewhere along the line, a person created an algo which then went on to mine data. These algos are generally heavily supervised, which means their outputs and regression outcomes are heavily censored by some expert to see that it makes sense. The skill of the expert remains paramount. Let these things run wild and you get comedy - albeit of a niche kind. Some of these are self optimising, but the structure of the algo is heavily checked and monitored.

Some people like to do fundamental analysis of the traditional kind. It can work. If none of it does, the capital markets are garbage.

Others see capital markets differently and like having heaps of data to hand (which is very different to saying that they actually use much of data at all in the actual implementation) which is used to correlate against subsequent stock price movements (these types of models are generally not fair value models, but relative return models and have time horizons for implementation of somewhat less than a year, typically). It too can work.

ML is just a tool, albeit a powerful one in the right environments (particularly sationary situations) and some situations in financial markets.

If you ran a simulation using a genetic algo to see how this all ends up, it's a seething mess that sees all sorts of aproaches having periods of success and failure. The market is organic. It's a journey.

If the markets were ruled by guys doing fundamental analysis with pocket protectors and pencils, and I was at good as they were, I'd bring a calculator to the gun fight and feel pretty good about it. Is that controversial? Naturally if a quant can't understand what it means to do those calculations in the first place, having that calculator allows you to pump out an amazing amount of total garbage.
 
tech/a said:
Yes I get that-----you see bots plugging away all the time.
But 2 things come to the Ducks Potato head.

(1) With such a broad base aren't results less likely to out perform and more likely to conform with the returns seen in an/the index?
Click to expand...

If you go long 1,000 stocks and short sell 1,000 other stocks, you get nothing like the index unless by an utter fluke of the highest order or purposefully doing it for a joke.

All else equal, there is nothing to indicate that the portfolio built this way has to perform better or worse than a concentrated one. You can always lever to an arbitrary risk level. What matters greatly is the risk/reward ratio.

Port con to a quant, to maximise this ratio, is yet another discussion. It's a heck of a lot more than whacking a stop loss on and putting 1% of portfolio at risk...although that approach is quantitatively sound under certain conditions for reasons I'll wager very few who use it actually understand.

tech/a said:
(2) In the case of a sudden move they still have the same problem getting out in a hurry.

This can be seen when funds take large hits like everyone else. Some of the smaller retail guys get out before and I know a few here have when they have seen/suspected it coming.
Click to expand...

What makes the use of ML imply that they all do the same thing? Do all T/A officiandos get hammered at the same time?
 
Yes I see your point.

Back to not caring which direction only that there is movement.
 
DaveDaGr8 said:
I would love to see AI beat a professional poker player and to be honest I'm not sure that it could. In poker, maths has very little to do with anything. Bluffing is everything. Good poker players can change the way the game is played to their advantage ie, get people to change their thinking. They could probably get an AI to modify it's thinking too right before they punish it.
.
Click to expand...

This is incorrect. High level poker is all maths, down to what hands to bluff with. You can prove mathematically how many hands you should be bluffing with and the ratio of bluffs to value bets in any given situation and you can prove mathematically what hands you should be bluffing with as long as your initial ranges leading up to that point is correct. The trouble with AI is that the game is so large and so complex that no computer has been able to solve it and currently the best human players can beat AI in no limit games.

The best players in the world like Ike Hakton, Doug Polk and OTB Red Barron play a highly mathematical game. OTB Red Baron is substantially better at no limit holdem then say Phil Ivy who is recognised as the best all round poker player in the world. However, he is certainly not the best no limit holdem player in the world. Phil Ivy plays the old school exploitative strategy where as OTB plays a highly mathematical game and crushes the competition.
 
ThingyMajiggy said:
Some interesting articles from Bloomberg:

https://www.bloomberg.com/news/arti...sustain-record-losses-in-setback-to-fink-plan

https://www.bloomberg.com/news/arti...ning-models-often-fail-to-learn-quicktake-q-a
Click to expand...

It seems like they are just trying to justify their management fees. Right now the average joe is probably thinking heh if I think the market is going up why don't I just stick my money in an index fund. If you put your money in an index fund after the GFC and just held it you would be laughing right to the bank right now. Hell, if you put your money in an index fund 20 years ago and just held through the tech crash and the GFC you would still be doing better than most people. I think that's exactly what the article is getting at, people just put their money in an ETF and be done with it. They pay very small management fee compared to hedge funds that probably won't even beat a market and they have little to think about.

I think a minimum standard for any system is whether it just just beat buying and holding an index fund forever. If you can't beat that by a significant margin why waste your time.
 
Valued said:
It seems like they are just trying to justify their management fees. Right now the average joe is probably thinking heh if I think the market is going up why don't I just stick my money in an index fund.
Click to expand...

Majority of average joes will never think that - as by their very definition average joe think they can outperform and in reality don't hence they stay average.

Valued said:
If you put your money in an index fund after the GFC and just held it you would be laughing right to the bank right now.
Click to expand...

Hindsight always easy..not many can time the end of GFC and also have the balls to enter.

Valued said:
I think a minimum standard for any system is whether it just just beat buying and holding an index fund forever. If you can't beat that by a significant margin why waste your time.
Click to expand...

Diversification and low draw downs are more important than absolute returns for high net worth portfolios.
 
minwa said:
Diversification and low draw downs are more important than absolute returns for high net worth portfolios.
Click to expand...

Well that's true because high networth individuals have to be concerned about asset allocation etc moreso then someone with a full time job who has money set aside for trading but the point is the same but it would only apply to the portion of the portfolio that is set aside for equities. Obviously, I am not suggesting someone go and take their portfolio which is say only 20% equities and then go sell all their property and go put it in an index fund. It applies to the portion of your portfolio that is set aside for trading in the equity markets. It equates to simply a question of whether you're beating the market while being in the market since you're in that market with a certain amount of your portfolio in the first place.
 
Valued said:
Well that's true because high networth individuals have to be concerned about asset allocation etc moreso then someone with a full time job who has money set aside for trading but the point is the same but it would only apply to the portion of the portfolio that is set aside for equities. Obviously, I am not suggesting someone go and take their portfolio which is say only 20% equities and then go sell all their property and go put it in an index fund. It applies to the portion of your portfolio that is set aside for trading in the equity markets. It equates to simply a question of whether you're beating the market while being in the market since you're in that market with a certain amount of your portfolio in the first place.
Click to expand...

I wasn't really talking about other assets, just the part of equities portfolio. A high net worth individual will pick a fund that has achieved a 30% gain with a drawdown of 15% over a index fund that may have achieved 40% gain with drawdown of 35% eg. like GFC type of periods for the majority of his portfolio. It's not wasting time to under return the index like you have put it if you are doing so with much lower risk. You only mentioned absolute return vs index while risk and draw down is actually what is more important for a fund because investors will sell out and reallocate when there is a big dip without participating in the recovery.
 
minwa said:
I wasn't really talking about other assets, just the part of equities portfolio. A high net worth individual will pick a fund that has achieved a 30% gain with a drawdown of 15% over a index fund that may have achieved 40% gain with drawdown of 35% eg. like GFC type of periods for the majority of his portfolio. It's not wasting time to under return the index like you have put it if you are doing so with much lower risk. You only mentioned absolute return vs index while risk and draw down is actually what is more important for a fund because investors will sell out and reallocate when there is a big dip without participating in the recovery.
Click to expand...

I think I am looking at this from a poker orientated aspect where you would always take the most profitable decisions regardless of the possible drawdown in your bankroll because the optimal way to play is to always take the highest EV option (the decision that will provide the most profit over the long term). Obviously, it depends on your life situation. If you're about to retire you definitely can't handle any drawdowns, you need that money. However, if you're young I would argue it is always correct to take the most profitable decision at any point. It depends on whether you can handle the drawdowns mentally though.
 
Valued said:
This is incorrect. High level poker is all maths, down to what hands to bluff with. You can prove mathematically how many hands you should be bluffing with and the ratio of bluffs to value bets in any given situation and you can prove mathematically what hands you should be bluffing with as long as your initial ranges leading up to that point is correct. The trouble with AI is that the game is so large and so complex that no computer has been able to solve it and currently the best human players can beat AI in no limit games.

The best players in the world like Ike Hakton, Doug Polk and OTB Red Barron play a highly mathematical game. OTB Red Baron is substantially better at no limit holdem then say Phil Ivy who is recognised as the best all round poker player in the world. However, he is certainly not the best no limit holdem player in the world. Phil Ivy plays the old school exploitative strategy where as OTB plays a highly mathematical game and crushes the competition.
Click to expand...
 
DaveDaGr8 said:
http://www.gizmodo.com.au/2017/01/w...man-poker-pros-are-getting-trounced-by-an-ai/
Click to expand...

That's pretty amazing if that's legit. I want to see a Libratus vs Libratus! Maybe the world will cave in on itself? :D

The way this stuff is advancing poses some questions about our future, they're talking there about how it's going to be used in fields like cybersecurity, medical diagnosis, in a couple years it looks like we won't be driving anymore(no delivery/truck drivers eventually?), factory jobs are out because of robots. I gather one day AI will be writing the actual code themselves too? What are we all going to be doing in the future? In what areas do humans have an advantage over AI? Creativity? On the spot reactions I guess, although in that article they're also talking about developing AGI, artificial general intelligence. Interesting times....especially to see how it develops in the trading world.
 

Similar threads

howardbandy
Getting Started in Machine Learning for Trading
2 3
Replies
50
Views
21K
entropy
E
jjbinks
Python - getting started
2 3
Replies
43
Views
19K
fiftyeight
fiftyeight
darkhorse70
The Beginner Trader's Manual Guide
Replies
9
Views
20K
commoyo
commoyo
Top