my journal 3

This is a discussion on my journal 3 within the Trading Journals forums, part of the Reception category; I am still trying to figure it out, the damn function used to calculate the combinations with repetition. I failed ...

Reply
 
LinkBack Thread Tools Search this Thread
Old Jan 31, 2012, 1:39pm   #185
 
Yamato's Avatar
Joined Mar 2003
user-defined functions and one more test with damn combo with rep.

Yamato started this thread I am still trying to figure it out, the damn function used to calculate the combinations with repetition. I failed again:
triangular_number_and_combinations_with_replacement.xlsm

But the nice thing is that this time I learned to quickly create user-defined functions, so it will be useful for the future. I created two of them. One for the triangular number and one for the damn function itself, which for a second I thought I could derive from the sum of all the triangular numbers from 5 to zero.

Pretty soon I'll give up and just be happy with having memorized the damn formula:
Click the image to open in full size.

What I wanted to do is focus on this thing, because I realize it's the foundation of probability and it's a neat little diagram, with the four possibilities, which in the future will help me categorize all probability problems.

Of course if I didn't have the job to worry about and to get up for, I'd sleep better and have more time for thinking clearly. Llike this I am kind of burned out all the time, but I am still somewhat moving forward.

Anyway, for today I learned something, I did some math (finished khan's review exercises), I did some probability, so I am ok.

I even resumed talking to my roommate, without exceeding into clowning. We talked for 15 minutes, and then I went back to working quietly. It would be perfect if we could do it like this every day, or twice a day.

[...]

Ok, one last try with the Italian wikipedia:
Calcolo combinatorio - Wikipedia

Quote:
Combinazioni con ripetizioni [modifica]

Quando l'ordine non Ť importante ma Ť possibile avere componenti ripetute si parla di combinazioni con ripetizione. Il numero di combinazioni con ripetizione di n oggetti di classe k Ť uguale a quello delle combinazioni senza ripetizione di n+k-1 oggetti di classe k ed Ť quindi uguale a:
Click the image to open in full size.. Ad esempio, vi sono Click the image to open in full size. modi di distribuire a 2 bambini distinguibili 4 caramelle indistinguibili, contando anche i casi in cui uno dei bambini non riceve nessuna caramella: 0-4, 1-3, 2-2, 3-1, 4-0. Equivalentemente, le combinazioni con ripetizioni informano sul numero di possibili n-ple di addendi non negativi la cui somma sia k (considerando diverse n-ple in cui eguali addendi compaiano in ordine differente); nel suddetto esempio, sono mostrate le cinque diverse duple di somma 4. Inoltre, le combinazioni con ripetizioni per n oggetti di classe k rappresentano il numero delle derivate parziali di ordine k calcolabili per una funzione a n variabili.
Ok, understood just a little more. The roommate is back, so I can't focus anymore. I might do some statistics now.

[...]

Yeah, he turned quiet again. I am doing it:
AP Statistics Tutorial: Transformations to Achieve Linearity

But before moving on to this, i found an interesting paradox or fact that I want to discuss because it's so counter-intuitive that it is worth mentioning.

If you have a coin, you have 50% of tossing a head. Now, instinctively, you'd say that if you toss it twice, you will then have a 100% chance of getting a head. That is close to the truth, with the law of large numbers, and if you toss it one million times.

But, interestingly, your chance of getting at least a head is only 1 - 0.5^2 = 0.75. In the same way you'd say that if your chance is 1%, by doing something 100 times, it will be 100%. And instead it turns out to be 1 - 0.99^100 = 63%.

That really strikes me as odd, but if you look at the tree diagram of coins, you see that it's the case:

snap1.jpg

You see? You toss the coin twice, but your chance is not 100% of getting at least one head. As you increase the number of tosses, the probability of getting at least one head increases (as a whole, not the individual ones), but it never goes to 100%.

Probability is a counter-intuitive field. Each time I have to do these examples over and over again, because by default I go back to the old thinking, that is wrong.

This is very related to trading, because let's say you're making a trade that you say is... 99% sure. You might make two opposite mistakes.

The first one is thinking that, as long as each time you make a 99% sure trade, you're ok. Indeed, after by making 100 such trades, as I said before you have a 63% chance of incurring one loss, so if you risk everything each time, then you'll blow out the account, with a probability of 63%, and a trader trades much more than that. Each time you have a feeling of a pretty much sure trade, and yet, by making many of them, your risk of failing once goes up each time. The same applies with other tiny risks we take each day, such as crossing the street, going out at night, and similar things. Each time you go out your chance of not being mugged goes down, exponentially (with a base a tiny bit lower than one).

The other potential mistake is to think that your probability is bigger than it is.

Anyway, it's interesting to think that if you go out at night and the risk of getting mugged is 0.01%, so your chance of not being mugged is 99.99%, which is 0.9999, then by going out every night for 10 years, your change becomes roughly 0.9999^3650=70%

I've done the stat trek material until minute 4.30 of the video, at the start of Flowchart.

This digression from... regression is anything but idle talk. I am in fact digesting the math I've covered in the past 4 months. I'd need a lot more digesting than this, but, since the final objective is math applied to trading systems, I am going at an acceptable speed. Of course I am not going to be teaching math after this. I will be happy if I can come out of this effort with the same knowledge of math as a good high-school student.

I was a bad high-school student, and I was particularly bad at... everything, and they made me stay back twice. I skipped months of classes. I used to go to the park and similar, or to watch movies. And I didn't do any homework. But Italian school will drive anyone intelligent crazy, because it's all about memorization and kissing up. I would have been better off without ever setting foot in a classroom. I would have learned much more from home schooling, but my parents didn't think of it. Actually the problem was precisely that my father did homeschool me, by giving me extra homework, because he said I wasn't getting enough at school - that's when the sadistic mother****er really screwed my academic future, because I got fed up with everything related to school and homework. Not only did I get to go to school and do that homework, but I got extra homework at home, and never once did he tell me "good job". Real ****ed up asshole I had for a father, and stupid mother who didn't have the strength and intelligence to stop him. What a ****ing bunch of idiots I had as parents. It really sucks.
__________________
Read: E.P. Chan, Cogneau - Hubner, Sewell, Tverberg. Search: expected shortfall, Monte Carlo VaR, extreme value theory. Trade.

Last edited by Yamato; Jan 31, 2012 at 3:10pm.
Yamato is offline   Reply With Quote
Old Jan 31, 2012, 6:45pm   #186
 
Yamato's Avatar
Joined Mar 2003
Re: my journal 3

Yamato started this thread Ok, I am back home. I finished the work on stat trek and got to this lesson:
AP Statistics Tutorial: Residuals, Outliers, and Influential Points

I will probably continue tomorrow because i need to digest this stuff. I am now 8 days/lessons away from this lesson:
AP Statistics Tutorial: Probability

which will finally reconnect me to probability. So it will be a little easier, and connected to reality and to what I've done recently.

Now I'll focus on the "combination with repetition" formula, for a change.

Jesus! Here's another one who came across mathsisfun and found their explanation of that formula to not be clear:
Yahoo! Canada Answers - How do combinations with repetition work?
Quote:
How do combinations with repetition work?

I need to learn about combinations and permutations. I get all both permutation variation (with repetition and without), I also get combinations without repetition. With repetition however is proving to be much more difficult. I used Combinations and Permutations to help me with the first three but the explanation there for the last one makes no sense to me. I tried to find another site to help me but they made even less sense.
The same exact experience I've had. And I also don't understand the explanation given but it is better than nothing.

I hope i'll have the strength to focus on this for the next few days.

__________________
Read: E.P. Chan, Cogneau - Hubner, Sewell, Tverberg. Search: expected shortfall, Monte Carlo VaR, extreme value theory. Trade.

Last edited by Yamato; Jan 31, 2012 at 8:33pm.
Yamato is offline   Reply With Quote
Old Jan 31, 2012, 7:34pm   #187
 
Yamato's Avatar
Joined Mar 2003
Re: my journal 3

Yamato started this thread Good link, with tons of categorized calculators:
Calculator Soup - Online Calculator
__________________
Read: E.P. Chan, Cogneau - Hubner, Sewell, Tverberg. Search: expected shortfall, Monte Carlo VaR, extreme value theory. Trade.

Last edited by Yamato; Jan 31, 2012 at 7:39pm.
Yamato is offline   Reply With Quote
Old Jan 31, 2012, 8:46pm   #188
 
Yamato's Avatar
Joined Mar 2003
still trying...

Yamato started this thread I haven't given up on combos and perms. Latest summary of the situation:
combos_and_perms.xls

In the meanwhile I ****ed up and went long on NG, a bit too early. I might lose all discretionary profit made until here. Just as expected. I had said I had run out of lucky trades.

[...]

I also doubled up when I noticed I was losing. Yeah, all the same old mistakes. I should just stay away from the computer. I knew this, too. It just keeps happening. Win 1, win 1, win 1, win 1, win 1, lose 5 or more (even the entire account).

I am done with math for the rest of the day. I ended up doing some of their kindergarten exercises, because they are awesome and there's no other exercises around:
Question

Now i'll start praying to make my trade go my way. It's down 600 now.

It either goes my way, break-even, or I can't close it. It's all psychological. That's why I can't trade discretionary.

Here's some good music, to encourage NG to rise:



Damn. It's putting it to sleep, that last part of the music. I am going to stop it and see how NG behaves. Of course I'll try to play other music to see what gets the best reaction.

Damn, I know what it is. It's the screaming child. We need to call the exorcist for the screaming child, or NG won't rise.

You know, I probably placed that damn trade because the child was frustrating me. Each time I hear that prick, I feel the urge to speed up my profit, in order to escape these neighbours. I think I will sacrifice the neighbours to NG, to see if it rises.
__________________
Read: E.P. Chan, Cogneau - Hubner, Sewell, Tverberg. Search: expected shortfall, Monte Carlo VaR, extreme value theory. Trade.

Last edited by Yamato; Jan 31, 2012 at 9:20pm.
Yamato is offline   Reply With Quote
Old Jan 31, 2012, 9:45pm   #189
 
Yamato's Avatar
Joined Mar 2003
Re: my journal 3

Yamato started this thread Interesting website:
ThatsMyFace - Custom Action Figures, Dolls & Masks from your Photos

Child is still screaming like he's possessed. NG still falling like it's possessed. I think it just hit bottom now, at -9%.

Nope, still falling. Child still screaming, too.

I am now losing 1500 dollars.

[...]

Instead of going to bed peacefully, with one NQ trade decided by the system, I am going to bed with a 1500 dollars ongoing loss, decided by me. I am sad. I did it. I can't blame anyone, except my personal personality, which still escapes me. I don't think I wanted to lose 1500 dollars. I was convinced I'd make 400 dollars. Then everything happened by itself.

I am staying long overnight, as there's high hopes for a recovery. One of the systems (a very good one) says so. I'll exit tomorrow morning. It might take my profit target during the night. Hopefully it will.

In that case, I will have missed only one automated trade, hopefully not a profitable one.
__________________
Read: E.P. Chan, Cogneau - Hubner, Sewell, Tverberg. Search: expected shortfall, Monte Carlo VaR, extreme value theory. Trade.

Last edited by Yamato; Jan 31, 2012 at 10:13pm.
Yamato is offline   Reply With Quote
Old Jan 31, 2012, 10:57pm   #190
 
Yamato's Avatar
Joined Mar 2003
Re: my journal 3

Yamato started this thread Hey, when I'll learn, I'll learn. I have similar problems to trading in my life.

1) In trading, I place trades I know will hurt my account in the long run.
2) In sleeping, in the morning i plan to sleep well the next night, and then, when the night comes, I go to sleep late
3) when eating, I plan not to eat, and then I eat

I've got some areas where I can't resist my urges, and do something that I know will hurt me. Deep inside, I knew that placing a trade wasn't a good idea. I've kept writing it. And yet I placed that one last trade, and now I am regretting it. And probably this wasn't the last discretionary trade I placed.

Probably tomorrow I'll pull it off and get out of it with no losses. In any case, it won't kill the account, but I still won't live to learn from it. Just like I never solved my sleeping problem, and I never was able to not eat if I am sitting at the table with food in front of me, or had beer in the refrigerator. I drink it immediately. And if I have cigarettes, I smoke them... I am like this.
__________________
Read: E.P. Chan, Cogneau - Hubner, Sewell, Tverberg. Search: expected shortfall, Monte Carlo VaR, extreme value theory. Trade.
Yamato is offline   Reply With Quote
Old Jan 31, 2012, 11:04pm   #191
 
Yamato's Avatar
Joined Mar 2003
Re: my journal 3

Yamato started this thread In many ways the market is like people, because the market is people. Check this video out, I posted it before:



The majority of people likes to gang up, and move in crowds. That's how a crowd is created. Because people prefer to stick together. So when there's a guy dancing, no one gets up. When there's an increasing number of people dancing, an increasing number gets up. Even those who weren't getting up before. So they just get up because there's a crowd doing it - not because they wanted to dance, but because they want to be with the crowd.

I am the opposite. Anyway, today they were ganging up on the NG. So the majority was dancing, while there were some people sitting on the grass. But this went on up to the close. Then the dip****s will go to bed.

Presumably, tonight, in the next few hours, there will be a different crowd, doing different things... and making the NG rise. This will be a contrarian-anticonformist crowd maybe, like me, but less impulsive. Then the other crowd will wake up and either follow what these guys have been doing during the night, or resume their previous behaviour. That's how you have this tendency of the market to go one way at certain hours and another way at other hours. Because the dip****s go to sleep. I am a special kind of dip****: I like to go against the other dip****s when they're an overwhelming majority, and say "hey, where the **** are you going...?", without completing my sentence because I get run over by the crowd. Instead of waiting for the dip****s to go to bed at about 10 pm CET, I anticipate them, and get run over.

It was just a mix of thoughts, not a perfectly coherent argument. The point of the whole argument is to stress once more how people suck.

And I suck, too. I'll learn when I'll learn. If I don't learn, I don't learn.

If I get away with taking irrational risks, I don't learn. If I don't get away, I still don't learn. I never managed to learn this thing, to not touch the trading. I always tried to prove that I could finally be smarter than my systems. And every 5 trades, I am reminded that I can't be, because I don't have their patience and memory of probability.

****, ****, ****.
__________________
Read: E.P. Chan, Cogneau - Hubner, Sewell, Tverberg. Search: expected shortfall, Monte Carlo VaR, extreme value theory. Trade.

Last edited by Yamato; Jan 31, 2012 at 11:16pm.
Yamato is offline   Reply With Quote
Old Feb 1, 2012, 6:39am   #192
 
Yamato's Avatar
Joined Mar 2003
Re: my journal 3

Yamato started this thread Damn. It keeps falling. Losing almost 2000.

At this point I am not exiting, because if I didn't exit earlier I can't exit now, and only a miracle can save me. Needless to say, all margin is used up by this ****ing trade I placed.

All the talking yesterday, about the crowd doing one thing, and others doing another thing... it was wishful thinking. Besides, every trade has two people involved, so you can't really say that someone is doing something. Well, you could say that a lot of people were pessimistic, were more willing to sell, than others to buy, and then price fell.

I don't have volume on tws, and I can't figure out why.

I am ****ed.

What made me do it?

Price has to reach that green line for me to break even:

snap1.jpg

I'm gonna go to work late. I'll sleep, until I can see this thing end, close at break-even...

Now it's going better:

3.jpg

After all it's less than one third bounce given all the falling it's been doing. I am only asking for it to bounce 25% of what it fell.

[...]

I've been doing some math. In the meanwhile, it's been doing a lot better:

5.jpg

Maybe I'll pull it off, one last time.

[...]

**** it. I am going to work on time, and I'll follow this thing from work. But I won't be able to close. I have my limit order in. It either takes it, or I am screwed.

I didn't sleep that bad, so i am going. Besides, I need to punish myself for making this trade, otherwise I won't remember it. Not that I ever remember these mistakes anyway, like going to bed on time and similar.

Actually now it's doing a little bit better, but I don't know how long it can keep rising like this.

8.jpg

I'm going to do some more math, finish the khan review exercises and the new ones (now the total is at 295, with ten questions apiece), and then I'll go to the damn place.

[...]

Ok, done with math and going to the office.

math.jpg

At least one chart is going my way.

I have a feeling, magical thinking, that NG will reward me for going to the office.

[...]

Back from shower. Still losing 1500.

You know, every body needs a doctor. That's why it's a good job. As long as there are people, there are bodies.

Yeah, I am going to work already. I am drying.

Neighbour bitch slamming her door. She always leaves an hour before me. I can't believe this stupid bitch still hasn't broken her leg or something. She's always wearing high heels, always running, and always comes back home at least once a day because she's forgotten something. This kind of stupid bitch should not have a high life expectancy. I can't believe she still didn't end up in the hospital. I just need a six months break.

Still losing 1500 and in the meanwhile two automated trades were missed, because margin was used up by the unprofitable discretionary trade, while the two discretionary trades are at break-even.

All right, I am going.
__________________
Read: E.P. Chan, Cogneau - Hubner, Sewell, Tverberg. Search: expected shortfall, Monte Carlo VaR, extreme value theory. Trade.

Last edited by Yamato; Feb 1, 2012 at 8:38am.
Yamato is offline   Reply With Quote
Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search

Similar Threads
Thread Thread Starter Forum Replies Last Post
my journal Fuzzybid Trading Journals 16 Jul 17, 2011 5:49pm
My Journal for all to see herronryan Trading Journals 44 Aug 26, 2010 4:32pm
A journal arabianights Trading Journals 66 Jan 10, 2010 2:09pm
My little journal. HaloTrader Trading Journals 16 Jun 26, 2009 12:02am

Currently Active Users Viewing This Thread: 9 (0 members and 9 guests)