Re: [amibroker] Re: Anyone actually making money?, AmiBroker Email List Archive

OMG this is so interesting to read you! I'll have to do it when I'll be more relaxed (after job... that damn job) and I hope I will be able to understand better what can or can not be done!

Just a fast question to Barry: you said the program takes 15-20 pages... Were you talking about a particular AFL you wrote (in this case we would be a long shot from the 12 lines Herman talked about) or were you talking about the manual to learn to code the AFL to be able to use it with IB automatic trading? I know I asked the question some time ago (well, wasn't exactly that question), but is it easy to set a simulation AT with IB?

To Dennis: I like your approach. I did the same: my gf stepbrother is a professional trader, been doing it for 20 years on smallcaps. He's watching his charts and everything all day long. I tried to do something that would be as close as possible to what he is doing and I was close to 30-40% AR on 15-minute chart. But... that's when I understood that there was a limit of 500 real-time symbols and that nearly killed the idea, because on small-caps you can never know in advance which stock will explode on one particular day. I tried to make the idea survive (there was a long thread with how to select the "best" 500 symbols) but it seems that does not do it. So this is why I am testing new ideas.

Again, I'll have to re-read everything that was written in this thread later (will probably comment again, or re-ask questions - you know me!). I never thought there would be so many interesting responses!

Thanks a lot again,

Louis

2008/7/31 Dennis Brown <see3d@xxxxxxxxxxx>

Louis,

I spent many years long ago writing my own rotational trading
simulation programs. I would trade by calling the broker and making
him sit on the phone with me while I waited for the perfect moment for
the entry or exit trigger. LOL That was when I negotiated a
fantastic rate of 5 cents a share. At the end of a year, I lost the
exact amount of my commissions. The losses were actually the result
of bad overnight announcements about once a month that dropped my
positions. I usually held only 3-5 stocks at a time, long only.

My simulations did not have a history of stocks that had sudden
losses, because I choose only 100 stocks that traded well in
simulations. The only problem was that all stocks have these
drawdowns at some point, but my simulations did not take that into
account. This is a well known mistake made by newbies. If I had the
internet at the time and contact with more experienced traders, I
could have saved a year and some losses in learning this simple
lesson. There are a lot of simple lessons that can be learned from
good traders or books written by them.

My current approach with AB is to take all the things I have learned
from very good discretionary day traders and try to figure out how to
automate much of that. I can trade that way myself, but I get
mentally tired after a few dozen ES trades in a day and start to make
stupid mistakes from a loss of focus. I look at AB as my trading
buddy that will keep me focused on the priority data, and present much
of the data I need in a way that I can digest in a glance. I also
program my AFL to talk to me, to let me know when a trade setup may be
about to happen --in case I get distracted.

I will continue to incorporate "MY" trading style into the evolving
AFL system as I see ways to describe what I do in algorithmic ways.
It is amazing how much pattern recognition the eyes/brain can do in a
second or less.

The morale to this story, is to first find a technical trading method
you like. Get to know it, understand why it would have an edge and
what its weaknesses are. Paper trade it by hand. Then, start to
program elements of it into AFL and test it. Examine when it works
and when it fails. Work with it until you feel it like second
nature. Make a few test trades with real funds to see how it works in
the real world, then go back to the drawing board again. There are a
lot more things to a successful trade than a good buy/sell signal.
Infrastructure issues can also have a noticeable effect. And as noted
the market moves in longer term cycles that affect strategies.

The process of finding "YOUR" trade from scratch and developing an AB
system should only take 3-5 years to consistent profits. A lot of the
time involved is personal training --"learning about yourself". Just
as in any skill or profession, it seems hard and like an impossible
job when you are learning. Then one day you wake up and suddenly you
just know how to do this without hardly even thinking about it. That
is the day you have become a trader, because it is in your nature.

For some specifics: I do not hold overnight. I trade off the 1-4
minute timeframe or equivalent volume or range bar charts. I only
trade futures on market indexes. I do not use standard indicators in
the way that you can find in the library. I mostly custom create my
own versions of indicators. Any indicators I use were modified based
on developing an insight about why they work the way they do. I only
take price and volume and time of day into consideration. I combine a
couple of systems with opposite characteristics into one system to
reduce drawdowns. My key metric is a consistently rising slope equity
curve with minimal drawdowns.

I have not completed all the details of my system to my satisfaction,
and I probably never will...

BR,
Dennis

On Jul 31, 2008, at 1:25 PM, bilbo0211 wrote:

> Hey Herman,
>
> I disagree with your comment about not learning from experienced
> traders. If that were true, why would a newbie listen to any advice an
> experienced trader like you has to offer?
>
> At the risk of offending you further, here is another quote from the
> trader with 20+ years of experience:
>
> "Using the same trading technique and exit strategy in all market
> conditions is likely to give you lots of practice at taking
> drawdowns."
>
> While the website I am quoting from (daisydogger) is a commercial site
> that sells trading information, there is some free information that
> should be valuable for beginners. In particular the article that
> describes a trading set up which includes how the trading parameters
> were optimized and the article on adaptive trading (how to adjust your
> trading to current market conditions). I wish I had that information
> when I started, it would have saved me a lot of time.
>
> Your suggestion that every newbie reinvent the wheel is likely to
> result in discovering 10,000 things that don't work before finding one
> that does.
>
> Bill
>
>
> --- In amibroker@xxxxxxxxxxxxxxx, "Herman" <psytek@xxx> wrote:
>>
>> I think it is a mistake for any beginner to try and imitate traders
> with 20+
>> years of, perhaps full time, market-related experience. These
> traders didn't
>> always trade the same way either, times change and so do the
>> markets. If
>> they are still trading they must have adapted many times. They
> probably have
>> developed skills that they can't communicate and that they
> themselves may
>> not even be aware of. Listening too carefull to anyone, without
> having the
>> same background, is going to cost you money. Big money if you take
> them too
>> serious. I speak from experience. Only if you do your own thing
>> will you
>> understand your actions and be a responsible trader. And, don't jump
> in with
>> big money. There is no shame in trading 1 share to prove a system.
>>
>> Take it easy, experiment, sit back and look at charts, try coding
>> some
>> simple systems. Remember: It's got to be fun! Trading system do not
> have to
>> be complicated. The best systems use simple techniques, with perhaps
> just a
>> dozen of AFL lines at most. Much the the coding complexities you see
>> discussed here have little or nothing to do with developing a trading
>> principle. Testing systems, even if they don't make money, will
>> reveal
>> things about the market that no amount of chart/tape reading can
> teach you.
>> Sooner or later you'll have a reportoire of simple ideas that you can
>> trade - that will be the time that you start posting on this list for
>> programming help :-)
>>
>> AmiBroker will not make money for you but it is (imo) the best
> development
>> tool around.
>>
>> Happy development,
>> herman
>>
>>
>>
>> -----Original Message-----
>> From: amibroker@xxxxxxxxxxxxxxx [mailto:amibroker@xxxxxxxxxxxxxxx]On
>> Behalf Of Paul Ho
>> Sent: July 31, 2008 9:49 AM
>> To: amibroker@xxxxxxxxxxxxxxx
>> Subject: [amibroker] Re: Anyone actually making money?
>>
>>
>> I'm still waiting for your to say something sensible, that's of
>> value,
>>> rather than just dig everyone out...???
>>
>> LOL, I just told you. YOu cant become a good trader until you know
>> yourself, really know yourself. Knowing how to draw a few lines makes
>> you a TA guy, not a trader. understanding the market makes you a
>> market analyst. Even know how to develop a system doesnt make you a
>> trader either, you are merely a system developer. When you master
>> yourself, you are on your way to become a real trader.
>> Just because you've been involved in trading for years, seeing others
>> making millions, no more makes your a trader than me an olympian just
>> because I've been watching olypmics for the last 45 years.
>> You might think I'm slagging you off, but in years to come, you will
>> find out for yourself whats true.
>>
>> --- In amibroker@xxxxxxxxxxxxxxx, "sidhartha70" <sidhartha70@>
>> wrote:
>>>
>>> Paul,
>>>
>>> I programme too... nothing wrong with programming and making your
>> life
>>> easier. And nothing wrong with asking for someone to 'throw you a
>>> bone' in that goal... I'm new to AFL. So what??
>>>
>>> I've been involved in trading for many years... market making, index
>>> arbitrage, statistical arbitrage, long/short prop etc.. many years.
>>> And some of the best money makers I've ever seen have been traders,
>>> not system traders, who genuinely have a great read of the market
>>> based off tape reading skills... they have achieved returns,
>>> consistently, that systems traders can only dream of...
>>>
>>> I've traded both... systems and otherwise... you can programme all
>> the
>>> most complicated indictaors you like... they are all based on the
>> same
>>> things tape reading is... which is price & volume action...
>>>
>>> I'm still waiting for your to say something sensible, that's of
>> value,
>>> rather than just dig everyone out...???
>>>
>>> --- In amibroker@xxxxxxxxxxxxxxx, "Paul Ho" <paul.tsho@> wrote:
>>>>
>>>> learn everything you can
>>>> about trendlines, support & resistance and the old tape reading
>> skills
>>>> and you will be miles ahead of 'system developers' in terms of
>>>> actually understanding the markets and becoming a better trader.
>>>> .......
>>>> So you think understanding a few lines on the chart you will
>> become
>>> a better
>>>> trader!!!
>>>> So you think understanding the market will make you a better
>> trader!!!
>>>> So you think you are miles ahead just because you know how to
>> draw a
>>> lines
>>>> rather than a system trader who write a few hundred lines of
>> codes.
>>>> NONE OF THE ABOVE WILL MAKE YOU A GOOD TRADER
>>>> YOU WILL ONLY BECOME A GOOD TRADER IF YOU KNOW YOUSELF.
>>>>
>>>> I'LL KEEP THIS IN A SAFE PLACE AND USE IT AS A REPLY WHEN YOU ASK
>>> SOMEONE TO
>>>> THROW YOU A BONE.
>>>>
>>>> ________________________________
>>>>
>>>> From: amibroker@xxxxxxxxxxxxxxx
>> [mailto:amibroker@xxxxxxxxxxxxxxx]
>>>> On Behalf Of sidhartha70
>>>> Sent: Thursday, 31 July 2008 10:47 PM
>>>> To: amibroker@xxxxxxxxxxxxxxx
>>>> Subject: [amibroker] Re: Anyone actually making money?
>>>>
>>>>
>>>>
>>>> Louis,
>>>>
>>>> Here's my tip... worth it's weight in gold. I've been doing
>> this for
>>>> a
>>>> long time...
>>>>
>>>> Stop looking for the holy grail system... learn everything
>> you can
>>>> about trendlines, support & resistance and the old tape
>> reading
>>>> skills
>>>> and you will be miles ahead of 'system developers' in terms of
>>>> actually understanding the markets and becoming a better
>> trader.
>>>>
>>>> Too many people spend far too much time trying to develop
>> a 'system'
>>>> rather than becoming a better trader.
>>>>
>>>> --- In amibroker@xxxxxxxxxxxxxxx
>>>> <mailto:amibroker%40yahoogroups.com> , "Louis P." <rockprog80@>
>> wrote:
>>>> >
>>>> > Hi,
>>>> >
>>>> > I was only wondering... Anyone actually making money or
>> making a
>>>> living
>>>> > with AB and trading?
>>>> >
>>>> > I've been working on ideas and plans for over 7 months now
>> and
>>>> didn't find
>>>> > anything convincing yet. I've been searching daily data,
>> then
>>>> hourly,
>>>> > 15-minute and now I am into 1-minute data and nothing seems
>>>> satisfying.
>>>> > Been searching RSI, MFI, ADX, MA, HHV, LLV... nothing seems
>> to
>>>> work.
>>>> >
>>>> > So... Anyone is making consistent money with this, and if
>> so, at
>>>> which
>>>> > timeframe and how do you do it?
>>>> >
>>>> > I'm beginning to think about switching to tick database; it
>> seems
>>>> even
>>>> > 1-minute is too slow for intraday trading. Anyone making
>> money
>>>> with
>>>> > 1-minute?
>>>> >
>>>> > Thanks,
>>>> >
>>>> > Louis
>>>> >
>>>>
>>>
>>
>>
>>
>> ------------------------------------
>>
>> Please note that this group is for discussion between users only.
>>
>> To get support from AmiBroker please send an e-mail directly to
>> SUPPORT {at} amibroker.com
>>
>> For NEW RELEASE ANNOUNCEMENTS and other news always check DEVLOG:
>> http://www.amibroker.com/devlog/
>>
>> For other support material please check also:
>> http://www.amibroker.com/support.html
>> Yahoo! Groups Links
>>
>
>
>
> ------------------------------------
>
> Please note that this group is for discussion between users only.
>
> To get support from AmiBroker please send an e-mail directly to
> SUPPORT {at} amibroker.com
>
> For NEW RELEASE ANNOUNCEMENTS and other news always check DEVLOG:
> http://www.amibroker.com/devlog/
>
> For other support material please check also:
> http://www.amibroker.com/support.html
> Yahoo! Groups Links
>
>
>