my journal 2

[Yamato]

Travis, I forgot to ask. What sort of forward testing did you do? Have you been running them in simulated trading?

With the automation, how did you do it in excel? Was it with the TWS API? Did you do it all yourself or are you using something from a code library?

I use the regular TWS platform from IB (not the paper trading account), but I have an excel sheet function that prevents the order from being sent to the broker, and yet records prices at the exact moment of receiving the signal. When I do trade them instead, I let the order go to the broker.

The automation is made via DDE Excel with TWS API, as you said. I did it all myself but only starting from 2006 (which means most of the work), because in 2005 a friend from another forum helped me get started - and without him this wouldn't have been possible. On the other hand, the systems we implemented were totally different from the ones I am using now, and they didn't work. That's why he abandoned the project after about one year, and I had to make all the changes myself. Eventually I got rid of those two systems that weren't working, after having created totally different systems. The "pioneer" systems I've spoken about in this post, were derived from those two original 2005 bid-ask systems. Despite having ended our little project, I felt the obligation to keep on sending him all the later updates of my systems up to less than a year ago. Right now, I really feel that this is all my work, and that I've shown enough gratitude.

 

[Yamato]

self-control update

I might start doing this on a weekly basis. Hopefully I won't forget doing it or it will be useless. As I said before, since I started doing this, I stopped scratching my head, slouching... and some other bad things I was doing. No effort needed. Simply monitoring and being aware of these things made me do them less.

 

[Yamato]

Airelon

Good friendly guy, explaining in simple terms some interesting things about futures:

I just don't like how he keeps on mentioning that the speculators are there to produce liquidity. He sounds like he's justifying what he's doing by saying he is useful to society or something. Why can't we live with the fact that the speculators are there to try and make money, regardless of what they will produce for society? I really don't understand why he has that point of view, so that in quite a few videos he keeps on saying "the speculators are there to produce liquidity". Who is he talking to? Just us, the speculators. Why does he keep on telling us that we're there to produce liquidity? I can't figure it out.

 

[episteme]

Not sure if you are happy with your technical platform. I will probably start trying to semi-automate some of my stuff in MetaTrader again when I have more time next month. If your ineterested in moving your stuff over to that platform let me know and I'd be happy to lend a hand.

 

[Yamato]

Yeah, thanks, but... I am totally happy with IB and their TWS. Actually I'd advise you and everyone else to use it. It's very light to automate everything on tws + excel. I can make quick changes... many advantages. Plus, since I can only program in VBA, I now have everything within my control, and don't have to rely on others. Yeah, thanks, but we can other types of project together. Like anything related to automated trading - no discretionary trading please because I quit, and haven't placed a trade in many days.

 

[episteme]

No problem. Maybe when I get the time i'll throw some ideas up on a forum here and see where we go.

 

[Adamus]

I use the regular TWS platform from IB (not the paper trading account), but I have an excel sheet function that prevents the order from being sent to the broker, and yet records prices at the exact moment of receiving the signal. When I do trade them instead, I let the order go to the broker.

ok so you're a VBA code monkey. I'm Java. Used to do VB a long time back and still have the O'Reilly VBA book on my shelf - thought I might need it some day to play around with spreadsheets. Amazing how similar Java and VB are. Just a few differences in syntax and all different names for functions and methods, but otherwise more or less the same when it comes to writing a bit of code.

I remember you throwing out your 'pioneer systems' - I bet it's sad to see 'em go, but you got to be cruel to be kind :jester:

You've done some mighty forward-testing. I'm interested to see how your systems pan out in real trading. If I remember correctly, it's about 1 trade per day, isn't it? Any idea what the average slippage is yet?

 

[Yamato]

I am a pretty basic programmer because I majored in political science and then, worked as a web designer for a few years (mostly html), and, in the last few years, I taught myself just enough vba to write these systems. I don't know anything about other programming languages (such as java), except if we can count easylanguage, which I needed to use for my back-testing on tradestation.

As I said a couple of posts ago, my systems trade 10 times per day on average. Or maybe a bit less. Like 9 or 8. No less than 8.

As far as slippage, I had written a post a while ago, which analyzed transaction costs, and listed these 4 categories:

1. commissions
2. bid-ask spread
3. slippage (delay in the execution usually gives you a worse execution price)
4. market impact (how your order will affect your own execution price)

If you agree with my definition of "slippage", I would say that my systems have no "slippage" at all, because of how fast the broker is and also because they make multi-hour long trades, so it's not an issue at all. Zero. Market impact is not an issue either. The transaction costs that include in my forward-testing are bid-ask spread and commissions.

 

[Yamato]

another great video by airelon

New video by airelon on futures' margins and lock limits in particular:

Lock-limits don't exist in every market and happen rarely on the markets I am trading, such as CL. But if they'll happen on CL (10 dollars), I'd be screwed. My systems are not prepared for such an event.

http://www.investorwords.com/5643/lock_limit.html
http://www.answers.com/topic/lock-limit
http://financial-dictionary.thefreedictionary.com/Lock+Limit

http://www.cmegroup.com/trading/equity-index/price-limit-guide.html

I have been going through the CME's web site for half an hour, searching for the lock-limit rules but except for the link above I found very little, unclear and too spread out. I am surprised, after hearing airelon's explanation of CL's lock-limit and it works, that I am not finding anything on CL's specifications page:
http://www.cmegroup.com/trading/energy/crude-oil/light-sweet-crude_contract_specifications.html

Ok, getting there....

I read this thread:
http://www.elitetrader.com/vb/showthread.php?s=&threadid=154759&highlight=lock+limit

Then went on Corn's specs page and found what they're talking about:
http://www.cmegroup.com/trading/commodities/grain-and-oilseed/corn_contract_specifications.html

It's called "Daily Price Limit".

Then I went back on CL's specs page and it is not there:
http://www.cmegroup.com/trading/energy/crude-oil/light-sweet-crude_contract_specifications.html

So what is airelon talking about? Could he be wrong? Or could the CME's CL page be wrong? I doubt it.

Now, this is good, because the ES does have a Daily Price Limit, too:
http://www.cmegroup.com/trading/equity-index/us-index/e-mini-sandp500_contract_specifications.html

So that could mean the CL doesn't have one.

Now GC doesn't have one either:
http://www.cmegroup.com/trading/metals/precious/gold_contract_specifications.html

Ok, found a thread on CL:
http://www.traderslaboratory.com/forums/f3/crude-oil-trading-limits-1563.html

Wow, what the heck...there's more on GOLD and OIL on other web sites than on CME's:
http://www.wikinvest.com/futures/Gold_Futures#Daily_Price_Limit
http://www.wikinvest.com/futures/Crude_Oil_Futures#Daily_Price_Limit

I've done enough work for now, but I would temporarily conclude that www.nymex.com is not a web site anymore, since the merger (or whatever it's called). And since that happened the Daily Price Limit table which they used to list nymex.com has disappeared from the web, and only stays in non-updated web sites, such as wikinvest.com and:
http://www.investopedia.com/university/commodities/commodities6.asp

This is where I'll start from when I'll continue my search:
http://www.google.com/search?hl=en&...p.com&btnG=Search&aq=f&aqi=&aql=&oq=&gs_rfai=

One more good link but which puzzles me even more, as the equity indexes are missing:
http://www.cmegroup.com/trading/Price-Limit-Update.html

Ok, I finally found the info for CL as well:
http://www.cmegroup.com/rulebook/NYMEX/2/200.pdf

200.06A Special Price Fluctuation Limits for Light Sweet Crude Oil Futures
(A) Initial Price Fluctuation Limits for All Contract Months. At the commencement of each trading day, there shall be price fluctuation limits in effect for each contract month of this futures contract of $10.00 per barrel above or below the previous day's settlement price for such contract month.

(B)
(1) Triggering Event and Temporary Trading Halt Related to Trading on Globex®. If a market for any of the first nine (9) contract months is bid or offered at the upper or lower price fluctuation limit, as applicable, on Globex it will be considered a Triggering Event which will halt trading for a five (5) minute period in all contract months of the CL futures contract, as well as all...

Way to go, CME... a goddamn labyrinth... thanks a lot for not writing anything on the CL specs page.

 

[Yamato]

self-control update

Quite high levels recently. Since I quit discretionary trading I think.

 

[Yamato]

I guess I spoke too soon:
http://www.trade2win.com/boards/trading-journals/85510-my-journal-2-a-52.html#post1081060

That **** sucker son of a bitch employee who has to give me the loan is making me wait forever. Maybe because we, as compliance, gave him a hard time for the past two years. Because he's a slacker and won't file the suspicious transaction reports on time and complete. Or maybe he just works like crap in every field, as far as granting loans and filing those reports. On the other hand, I did get my previous loans within 24 hours, so I don't know what to think.

On the other hand, it might be good that they make me wait for a week, so I can stop being restless, as it's happening indeed. They're forcing me to take a break. Ever since I stopped discretionary trading I've stopped stressing out, rushing home to watch my computer... we'll see. Of course if I don't get my loan within a week, I will call the lady and ask her if there's a problem. Of course the lady is a nice one, but the guy whose signature she has to get is that asshole I was writing about.

 

[Yamato]

feedback on my journals

Over 500 posts of feedback so far on my 2 journals, from 74 different people.



Only a few of them posted on both journals, precisely 7. That's surprising. As time goes by that number will increase, since the new journal is only 500 posts long.



This is the file, in case anyone is interested:

View attachment who_posted_20100319.xls


Another interesting thing I found is that my percentage of posts on both journals, despite them being written at different times and having different users in them, is always slightly above 80%.



Also, comparing my journal to others, I am noticing that I attract very few (quality) readers and even fewer who post of course. Another thing is that I've benefited from what they post. So, once again, thank you, readers, for suggesting me things to read, change, test. I rarely listen, but when I listen, it always helped me. Like the famous excel formula for calculating maximum drawdown I developed with Weighbridge, which to date is the single greatest contribution I've received for my automated systems:
http://www.trade2win.com/boards/trading-journals/72598-my-journal-44.html#post953574

Believe it or not, I used to manually calculate the maximum drawdown, which is necessary for automated trading, by summing up the losses of each and every one of my 40 systems, on a weekly basis. By now I'd have given up, but advised by and with Weighbridge we developed that formula you can find at the post linked above.

 

[Yamato]

automated trading update

Pretty good.

Something I noticed is that, in order to decide whether signals need to be reversed, I've been measuring the profitability/unprofitability of the past week on friday, but there's entries on friday night which can only be computed on monday. So this could screw up my trading somehow. The most proper way to measure therefore would be to measure the past week's profitability both on friday night and on monday night. But it's too complicated, and the difference is not enough to reverse a week's profitability usually. This past monday turned friday's weekly gains from 500 to 300, and it was an exception in many ways.

Indeed, another thing to mention is that this week we barely did not reverse signals, with just a profit of 300. So this week the above-mentioned change between friday and monday could have made a difference. And I could have missed reversing signals that should have been reversed.

Another interesting thing is that I have not been making a lot: 300 last week, and 1200 this week. But it's ok because I didn't lose.

 

[Yamato]

self-control update: the power of using a system and accepting drawdowns

An unimaginable achievement: I stopped scratching my head for 3 weeks
Even if I didn't do anything else, I achieved this already: I have not scratched my head for about 3 weeks, a problem which I thought was impossible to defeat. Before now, it hadn't happened for years.


Like for trading systems, it's not "all or nothing"
The trick was this: you shouldn't consider your vices/addictions as something to defeat all at once, but as something to cut back on, as much as possible. Because otherwise if you say "from now on, never again", the minute you fail, you just fail all the way. You want to be able to allow yourself some leash, and be able to partially fail objectives, so you can still work on them, rather than fail for good, and give up on the project altogether. Until now, before this great idea of the excel sheet, I used the wrong method, "from now on, never again", I failed within a few days, gave up for another two years, then started again... and since I failed so much I started thinking I could never solve the problem. The trick is to be a little more tolerant towards yourself. For example: I might be extremely stressed out two weeks from now, and I might scratch my head for a whole day, but that's just one day on my chart, and I've got two months behind me of good results.

Drawdowns happen in life as in trading
It's like a drawdown for trading. One bad day, one loss, doesn't make your system defeated forever nor worthless. Ever since i started monitoring on a table and chart, and looking at my fight towards negative addictions as the equity line of a trading system, I also started being tolerant towards myself and accepting a drawdown, and that's when I learned to stick to what is a profitable system to fight my addictions (the system is simply to monitor them). Just like with a trading system, you can't expect your system to win every single time. Only a chart and a table enable you to see that, and to remember you have progress and good results behind you, and that a bad day doesn't make your method worthless.

 

[Adamus]

As far as slippage, I had written a post a while ago, which analyzed transaction costs, and listed these 4 categories:

1. commissions
2. bid-ask spread
3. slippage (delay in the execution usually gives you a worse execution price)
4. market impact (how your order will affect your own execution price)

If you agree with my definition of "slippage", I would say that my systems have no "slippage" at all, because of how fast the broker is and also because they make multi-hour long trades, so it's not an issue at all. Zero. Market impact is not an issue either. The transaction costs that include in my forward-testing are bid-ask spread and commissions.

I've already asked if I can have your system, haven't I?

No slippage. :|

I don't want to get all pessimistic on you, but are you sure about that? If I get one tick slippage more on each trade than I estimate, it breaks the high frequency systems I am looking at. Yours must be different, or you're going to find out your wrong. Are you entering on limits, at mkt or on stops?

 

[Yamato]

Wow, thanks, I am flattered. As I said, if you have any business proposals, I am all ears. I wish I could find ways to raise some capital, but so far I haven't figured out any ways other than trying to trade them myself (with no success so far, due to the problems I wrote about). So I took that mother ****ing loan, which hasn't come through yet, thanks to that mother ****er lazy employee, who still hasn't signed my papers. Because he's a slacker.

As far as slippage, I've said enough about why they don't have it (provided we agree on the concept of "slippage", cfr.post above): in short, they're not high-frequency systems like yours. The only way they'll stop being profitable is if they stop working (e.g.: the markets change, or similar), and not because I underestimated their transaction costs. I am pretty sure about this.

I mean: I know my many problems (discretionary trading, interfering with systems, etc.) but I also know what I am doing right. Besides, we're talking about forward-testing results (with real prices, in real-time, taken straight out of TWS), so most of the potential miscalculations are already out of the picture. With back-testing results I would have to be more careful.

Besides, I can give you advice on how to develop your own systems, in case we can't find any viable business arrangements. One thing is: do not try to make high-frequency systems. Unless of course you already found some that work. I was unable to find high-frequency systems that were profitable. The longer the timeframe, the better. Precisely because the longer is your timeframe the lower are your transaction costs. Imagine if you were making trades that last one week on the CL, and had a target profit and stoploss of several thousand dollars. How relevant would your transaction costs be? Totally irrelevant.

Regarding your question about my entries: I am entering at MKT, always. Same for my exits.

 

[FXTrend240]

Re: feedback on my journals

Over 500 posts of feedback so far on my 2 journals, from 74 different people.

View attachment 79048

Only a few of them posted on both journals, precisely 7. That's surprising. As time goes by that number will increase, since the new journal is only 500 posts long.

View attachment 79050

This is the file, in case anyone is interested:

View attachment 79052


Another interesting thing I found is that my percentage of posts on both journals, despite them being written at different times and having different users in them, is always slightly above 80%.

View attachment 79054

Also, comparing my journal to others, I am noticing that I attract very few (quality) readers and even fewer who post of course. Another thing is that I've benefited from what they post. So, once again, thank you, readers, for suggesting me things to read, change, test. I rarely listen, but when I listen, it always helped me. Like the famous excel formula for calculating maximum drawdown I developed with Weighbridge, which to date is the single greatest contribution I've received for my automated systems:
http://www.trade2win.com/boards/trading-journals/72598-my-journal-44.html#post953574

Believe it or not, I used to manually calculate the maximum drawdown, which is necessary for automated trading, by summing up the losses of each and every one of my 40 systems, on a weekly basis. By now I'd have given up, but advised by and with Weighbridge we developed that formula you can find at the post linked above.

i posted on both guess who

 

[Yamato]

Yeah, quite balanced posting: 10 on "my journal" and 17 here. I guess you are the person who posted most equally between the two journals, if we only consider people with many posts. I found out who you are by looking for, among the frequently-posting users on "my journal", those who were banned, and then I double-checked with the favorite books.

 

[Yamato]

astrud gilberto

Check out how beautiful these songs are. They are very well made. I don't care if they're not hers, and if she's not as talented as her former husband. They're very well made. I would say the best bossa nova ever sung.

http://en.wikipedia.org/wiki/Astrud_Gilberto
http://www.amazon.com/Compact-Jazz-Astrud-Gilberto/dp/B00000472W

 

[Yamato]

best music channel

This guy is awesome, brazilian:
http://www.youtube.com/user/luansk88

He's got the best collection of youtube favorites and most of them can be posted (no protection and stuff):