Automated Trading - Is there hope?

[Liquid validity]

Just that I intend to retire from trading after a few more years - I've had enough of sleepless nights. I've had winning streaks of over 30 working days in a row and I STILL can't get more than 4 hours during those times. Even less during others.

The idea that you've just been found out and been lucky for the past x years never goes away on a losing day... the lows of a losing day are much lower than the highs of a winning day, no matter the magnitude of the latter vs the insignificance of the former.

In my mind, anyway.

Why don't you put some cash into generating rental income and downscale the trading?
Thats what I would do, best yields aren't London, unless asset appreciation
is what you want with rent as a dividend.
Even with foreign inflows of capital into prime London real estate, it does
have bubble written all over it...

 

[Liquid validity]

I wouldn't wish automation on anyone. It's like a nasty recurring disease that you spend 2 years of your life trying to cure and then it still flares up in the subsequent years.

If anyone thinks they can automate without working harder than they have ever worked and giving up all their social time and weekends for a few years then good luck to them.

I'm still amortising my lost time each time I return!

Somehow I missed this reply before, don't know how, time for an eye test :-0
Completely agree, most people aren't aware of the inefficiencies that come with
automation, binary logic and granularity are the 2 biggest factors that have
caused me the most difficulty.

Completely with you on the social time and spare time, same here.
TBH, at the time, I actually wanted to do it, even though I was aware of the cost.
Even now, yeah there is always a price to pay, a sacrifice to make of some sort,
same with anything really.
For me personally, I wouldn't change a thing, and that has no relation to
any success or lack of, its simply something I was driven to do, regardless of success or failure.

 

[JLM Financial]

I believe EA's and automated trading systems are dangerous. Sure, they may work for short periods of time. However, in the long term they will fail unless they are continuously tweaked to adapt to the changing markets.

As a couple of people have mentioned, all that time spent learning to code and actually coding could be spent elsewhere. That being said, there may be a few who have found EA's and automated trading systems profitable but for the majority, it is unlikely.

 

[Pat494]

I was thinking about trading in general and now along comes a struggling Blackberry. Not good enough to compete with Apple/Samsung etc. but could have a future being programmed up for trading if it was profitable. Perfect for black box systems too..

The talking heads reckon Blackberry although nominally worth billions will wither away as it is unable to increase market share and has already tried re-marketing itself at some cost. The tech world moves so rapidly, this once mighty company could be next to worthless sooner rather than later. The Blackberry quotes for Sept/Dec spreads are closed currently !!

 

[Liquid validity]

I was thinking about trading in general and now along comes a struggling Blackberry. Not good enough to compete with Apple/Samsung etc. but could have a future being programmed up for trading if it was profitable. Perfect for black box systems too..

The talking heads reckon Blackberry although nominally worth billions will wither away as it is unable to increase market share and has already tried re-marketing itself at some cost. The tech world moves so rapidly, this once mighty company could be next to worthless sooner rather than later. The Blackberry quotes for Sept/Dec spreads are closed currently !!

You've lost me, maybe you can elaborate

 

[Pat494]

You've lost me, maybe you can elaborate

err not sure where I lost you LV but seeing as to how we are all interested in making money I thought the above observations might yield a few bob ?

 

[Liquid validity]

err not sure where I lost you LV but seeing as to how we are all interested in making money I thought the above observations might yield a few bob ?

The bit about programming a blackberry for trading...
Anyway, I don't trade U.S. stocks, so I can't comment.

 

[random12345]

Why don't you put some cash into generating rental income and downscale the trading?
Thats what I would do, best yields aren't London, unless asset appreciation
is what you want with rent as a dividend.
Even with foreign inflows of capital into prime London real estate, it does
have bubble written all over it...

Yeah you might be right - I only have a small proportion of cash tied up in trading, but once you hit a certain £ point of losing trade (I guess this depends how rich you were as a kid - I was not!), the psychology doesn't really change.

My only 'intelligent' investment in B&M has been commercial property in the latest w@nker, sorry 'hipster' areas - I bought commercial properties in Hoxton that have appreciated quite a bit and I much prefer to deal with business owners when it's in their interest to keep the place clean and fully repaired and insured.

I don't want the headache of being a residential landlord. I know a few guys who have no end of trouble even with a management company in between.

 

[random12345]

Somehow I missed this reply before, don't know how, time for an eye test :-0
Completely agree, most people aren't aware of the inefficiencies that come with
automation, binary logic and granularity are the 2 biggest factors that have
caused me the most difficulty.

Completely with you on the social time and spare time, same here.
TBH, at the time, I actually wanted to do it, even though I was aware of the cost.
Even now, yeah there is always a price to pay, a sacrifice to make of some sort,
same with anything really.
For me personally, I wouldn't change a thing, and that has no relation to
any success or lack of, its simply something I was driven to do, regardless of success or failure.

Yeah I agree - if anything else treated you in this way, you would give in, but there is always a driving force and I don't think it is just money. More a desire for success in some field, especially if you couldn't get any satisfaction out of a 9 to 5.

I always have to hide my anger when a friend starts going on about some failed venture that they put all of 2 months effort into!

 

[random12345]

I believe EA's and automated trading systems are dangerous. Sure, they may work for short periods of time. However, in the long term they will fail unless they are continuously tweaked to adapt to the changing markets.

As a couple of people have mentioned, all that time spent learning to code and actually coding could be spent elsewhere. That being said, there may be a few who have found EA's and automated trading systems profitable but for the majority, it is unlikely.

There is truth in this - only those with dangerous levels of paranoia regarding successful systems can really make it in automation - you have to be willing to reassess almost every weekend.

The most difficult part is in the confirmation of your core concept - walk forward optimisation isn't an answer and nor is forward sampling. You need a relevant sample for recent market shifts without being excessively reliant on curve fitting and then you need to take it live to know it's a feasible going concern at that sample size, you can't know until you try it. So you're looking at around a year of live testing (with returns as expected per your back-testing metrics) with numerous revisits to the drawing board, plus development time around the issues that come with live like bad fills, suspensions and the like.

This is why I always get nervous by people saying neural nets work for them, but they can't explain why it works, especially when these nets leave losers open and the owner is unsure when they will be closed, even if hedged. For me, I can explain why, how and give a good indication of expected monthly return.

 

[Pat494]

I was thinking about trading in general and now along comes a struggling Blackberry. Not good enough to compete with Apple/Samsung etc. but could have a future being programmed up for trading if it was profitable. Perfect for black box systems too..

The talking heads reckon Blackberry although nominally worth billions will wither away as it is unable to increase market share and has already tried re-marketing itself at some cost. The tech world moves so rapidly, this once mighty company could be next to worthless sooner rather than later. The Blackberry quotes for Sept/Dec spreads are closed currently !!

The quotes for Blackberry are now available.
I will be interested to see what has happened in 6 months from now !!

 

[new_trader]

There is truth in this - only those with dangerous levels of paranoia regarding successful methodologies can really make it in trading - you have to be willing to reassess almost every weekend.

The most difficult part is in the confirmation of your core concept - walk forward optimisation isn't an answer and nor is forward sampling. You need a relevant sample for recent market shifts without being excessively reliant on curve fitting and then you need to take it live to know it's a feasible going concern at that sample size, you can't know until you try it. So you're looking at around a year of live testing (with returns as expected per your back-testing metrics) with numerous revisits to the drawing board, plus development time around the issues that come with live like bad fills, suspensions and the like.

So why automate?

 

[Liquid validity]

Yeah you might be right - I only have a small proportion of cash tied up in trading, but once you hit a certain £ point of losing trade (I guess this depends how rich you were as a kid - I was not!), the psychology doesn't really change.

My only 'intelligent' investment in B&M has been commercial property in the latest w@nker, sorry 'hipster' areas - I bought commercial properties in Hoxton that have appreciated quite a bit and I much prefer to deal with business owners when it's in their interest to keep the place clean and fully repaired and insured.

I don't want the headache of being a residential landlord. I know a few guys who have no end of trouble even with a management company in between.

Fair points on residential, sounds like another source of income won't
be a problem anyway.
Funny you mention childhood wealth, seems to be pretty common factor (lack of),
that creates the drive in certain people, not always, its certainly been
a pretty common theme in biogs I've read etc.

 

[Liquid validity]

So why automate?

Well I can't speak for anyone else, but me personally,
it was mostly an availability issue.
I simply was unable to be present to discretionary trade
the prime time of the UK and USA sessions.
Thats it.

For me it has never been about efficiency, someone who knows what
they are doing could beat my automated returns with discretionary trading.

 

[SlowlyButSurely]

"My only 'intelligent' investment in B&M has been commercial property in the latest w@nker, sorry 'hipster' areas - I bought commercial properties in Hoxton that have appreciated quite a bit and I much prefer to deal with business owners when it's in their interest to keep the place clean and fully repaired and insured."

It's amazing how much East London has changed in the last 5-10 years. A great investment.

 

[jack.r]

I would say automata trading is good, but only if you created the given EAs. I made huge losses, while testing different EAs, but I still think we have to find the balance between automata and manual trading. Nowadays, I don't really have time for trading, therefore I decided to invest a higher amount at a professional and regulated European trader, and I can say it was a good decision. Since January he already made 54% profit on my trading account, so i am more than satisfied. Although I know when I will have time, I will continue trading, as I love trading on forex.

 

[random12345]

So why automate?

Few reasons:

- More measurable returns as regards daily SD vs a historical sample/variance personally help me with psychology as I don't have the motivation to build up a manual sample.

- I trade 21 hours a day.

- I don't think I would be able to outperform them manually unless I was a trading savant. I am not Paul Rotter.

- I have a poor boredom threshold regarding waiting for entries.

- As with LV I wrote my own algos primarily because I was otherwise engaged during the day.

Having said that I have just lumped some cash into RCG to blow some cash on the FGBL. "Hello old friend."

 

[random12345]

"My only 'intelligent' investment in B&M has been commercial property in the latest w@nker, sorry 'hipster' areas - I bought commercial properties in Hoxton that have appreciated quite a bit and I much prefer to deal with business owners when it's in their interest to keep the place clean and fully repaired and insured."

It's amazing how much East London has changed in the last 5-10 years. A great investment.

Yes it is ironic to say the least to see middle class Uni kids enjoying the latest crappelatte in Hoxton Sq.

 

[fozzybear]

(I'm horrible at writing my thoughts down, so here goes)

What I don't understand, is that successful traders say "come up with a plan, write it down, and stick to it".

So, if you do this, and it's a winning plan, you should be able to wrote code to implement it.

If you can write code to implement it, it's an automated trading.

A lot of successful traders never have backtested their system, yet it works for them (I'm not trying to take anything away from them), yet they bash automated trading.

I'm having a hard time understanding how someone can bash an statistically valid system, but yet have never backtested theirs.

Does that make sense?

--fozzy.
PS:Not trying to step on anyone's toes here, just trying to understand the mentality.

 

[Liquid validity]

(I'm horrible at writing my thoughts down, so here goes)

What I don't understand, is that successful traders say "come up with a plan, write it down, and stick to it".

So, if you do this, and it's a winning plan, you should be able to wrote code to implement it.

If you can write code to implement it, it's an automated trading.

A lot of successful traders never have backtested their system, yet it works for them (I'm not trying to take anything away from them), yet they bash automated trading.

I'm having a hard time understanding how someone can bash an statistically valid system, but yet have never backtested theirs.

Different strokes for different folks.
Both can fail.

Discretionary trading subscribes to the theory that every moment is unique and unrepeatable (which is true).
The aim is to trade what is happening, rather than try and fit a trade to a set of broad and rigid
rules.

Automated trading (that works for a reasonable length of time) accepts that
each moment is unique, but there is also a certain common behaviour in markets
that take a long time to change.
Automation aims to profit from this common behaviour, and accepts the
inefficiencies that arise from unique moments that don't conform to that general
behaviour.
In practice that can mean automation is more susceptible to prolonged periods
of making little or no money, or even quite deep drawdown.

Its all shades of grey, retail automation is generally not as efficient as retail discretionary.
That is assuming a sound proven methodology, without that, its straight to the poor house.
Institutional automation can in certain cases wipe the floor with everything .
In other cases, dramatic blowups result.

Neither way is any guarantee.
Anyone and anything can fail or go into drawdown.
Potential failure and drawdown are probably the most
important things to focus on, no matter how you trade.