my journal 3

Another very interesting interview with assad

Charlie Rose | Bashar al-Assad - President of Syria - YouTube

Dude, this guy is more of a dictator than obama, and syria is more of a dictatorship than the US, for sure. But this man lies less than obama, maybe precisely because he's a dictator. He doesn't have to please people.

I had never heard such interesting interviews from politicians. I guess dictators are the most interesting interviewees.

His wife, he met her in london while studying ophthalmology (she's syrian, but born in london):

Asma al-Assad

At least from just 10 minutes listening to her, she doesn't seem as intelligent as her husband. To tell you the truth, she doesn't sound smart at all, like the typical dumb college chick. She's smiling and laughing too much. Dude, you're the wife of a goddamn dictator: you're not allowed to laugh and smile. I mean, this is serious stuff. Your husband is giving orders to kill people, so I am still against the US overthrowing him as it's trying to do, but being the wife of a dictator, you just simply shouldn't laugh or smile and pretend everything is fine.
 
Last edited:
‫

What is interesting to me is that both the rebels and the syrian arab army (the government) shout "allahu akbar", whenever they open their mouth. I don't think i see anything similar in the west.

rebels freeing prisoners:

 
Last edited:
Unbelievable. Another unprofitable week by my systems. This is the third unprofitable week in a row, and if we count the last 3 months, my systems have made zero money. This is really testing me.

At the same time, my discretionary trading seems to be improving, in that I am defeating my pride and perfectionism (not expecting to get every trade right).

How can it be that when I am not looking, my systems make all the money, and then, when I trade them, they make almost no money at all?

I'm going to have to wait another two weeks to find out. So far this really sucks. Instead of making the habitual 50k per semester, this last semester they're going to make 10k or so. Have been trading them for about 20 weeks so far.

Let's wait another week... and then another...

... either the markets have momentarily changed, or my systems have stopped working... i can't believe this, after their having worked for years, they can't stop working exactly when i start trading them.

capital now slightly above 14k.
 
Last edited:
Tunisian Ex-Jihadist Tells Tunisian TV Why he Stopped Fighting Alongside the FSA in Syria - YouTube

Ali Garboussi, he went to fight in syria as a jihadist, then after a month, he changed his mind and returned to tunisia.

http://en.wikipedia.org/wiki/Proxy_war
A proxy war or proxy warfare is a war that results when opposing powers use third parties as substitutes for fighting each other directly. While powers have sometimes used governments as proxies, violent non-state actors, mercenaries, or other third parties are more often employed. It is hoped that these groups can strike an opponent without leading to full-scale war.
Proxy wars have also been fought alongside full-scale conflicts. It is almost impossible to have a pure proxy war, as the groups fighting for a certain nation usually have their own interests, which can diverge from those of their patron.
Click to expand...
 
Last edited:
travis said:
Unbelievable. Another unprofitable week by my systems. This is the third unprofitable week in a row, and if we count the last 3 months, my systems have made zero money. This is really testing me.
Click to expand...

Do your systems account for a different trading dynamics due to the holiday season?

One example: more trading is done during the start of a trading session and the A team doesn't return to the trading desk after their lunch break (they are probably busy buying gifts or going to the end of year parties).
 
By all means do they exploit cyclical and seasonal patterns, given that they all have time entry and exits, and also they behave differently according to the day of the week, and sometimes even month of the year.

However, each follows its own pattern, and I am trading 26 different systems on 12 different futures, so their seasonal biases cancel each other out.

I also checked and there's no difference in profit according to months.

But the trend that I see is that indeed, when I start trading them, they don't perform as well as when I picked them. And to a certain extent this is normal, because if you pick the winning team and expect it to win the next tournament, sometimes you will not be right, even though past performance has a predictive advantage that should be used.

At the same time a user has suggested a doubt to me, in a private message, and that is: what if the systems' performance constantly declined due to various reasons?

And, if things keep going like this, this might be a good additional explanation.

...

Another thing I should add is that my systems perform better when the markets either go sideways or rise. I am not saying they lose when the markets go down, but they don't do as well.

I am going to post a chart showing this, in a few minutes. I have to create it first.

...

Here it is, finally:

26_systems_vs_EUR_vs_SP500.JPG

They seem indifferent to the EUR/USD, and it seems that when SP falls, they break even or lose very little. Then, if markets rise (if SP rises, and the correlated markets), they do very well. So I guess this was useful because it made me notice that the markets have been falling a little bit lately, so that could partially explain it. In other words, if the week has been negative for SP500, I shouldn't be expecting to make money.

...

For my own record, the 26 systems I am now trading (and described in the chart above) are the following:

AUD_ID_01
CAD_ID_01
CAD_ID_03
CHF_ID_01
CL_ID_04
CL_ID_05
CL_ON_03
EUR_ID_05
GBL_ID_01
GBL_ID_02
GBP_ID_02
GBP_ID_04
HG_ID_03
HG_ID_07
NG_ID_01
NG_ID_02
NG_ID_04
NG_ON_02
NG_ON_04
NQ_ID_02
NQ_ID_03
NQ_ON_01
NQ_ON_02
YM_ID_03
YM_ID_05
ZN_ON_02
 
Last edited:
travis said:
what if the systems' performance constantly declined due to various reasons?
Click to expand...

Many times systems are overoptimized during development. This happens even if we are aware of the problem and try to avoid it. Such a system will not perform as in the past.

One simple approach would be to test and develop on one set of data. When you think you are done, test the system on another set of data. If performance is similar, keep it. If performance differs significantly, discard the system (meaning stop developing it). If you go back to change some parameters and test again on the second set of data you already over-optimized the system (now you are curve-fitting it to both sets of data). The important thing here is that you only test it once on the second set of data.
 
Yes, I agree. I call it "out-of-sample testing". I've been doing it. And on top of that, I've been doing forward-testing, for years.
 
You must log in or register to reply here.

Similar threads

Yamato
  • Locked
my journal 2
169 170 171
Replies
3K
Views
476K
timsk
timsk
J
  • Poll
Trading journals: general discussion
Replies
5
Views
2K
athvia
A
N
Ninja's Journal
Replies
5
Views
2K
Ninja377
N
G
need help with Journal template and macros
Replies
0
Views
2K
grego99
G
T
Trading Journals and No Trading Journals
Replies
14
Views
3K
hhiusa
hhiusa
Top