Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
Under the auspices of our leader and to personally curb my enthusiasm to jump into the waters of discussion, i decided to take the best advice and put in a pseudo journal here.
First an introduction to give things context:
Born as a poor child in the Bronx to a father who survived WWII in Displaced persons camp, and a mother born here as a child of parents, my life started enrolled in the school of hard knocks (along with my younger sister). No reason to go through the litany of the decades from the 1970s onward (though i was born earlier, I was then too young to be a part of things), suffice it to say, life in bad places is seldom, except in movies, felt as bad (unless you want to believe your a victim of circumstance, which we all are). I worked hard, made it into Bronx High school of science, and come college, feminism hit its peak and forbid me the cash to become a scientist/researcher. Sis though feeling she earned it, has many more degrees because of the largess of the ages. I on the other hand, sufficed with certificate courses, sometimes living on the street in NY to attend, and eked my way through to start a life in computer science. As is the wont of the young and foolish, i made some not great friends, lacked wisdom, and eventually got involved with a woman that could easily have inspired and expanded on the 'gone girl' repertoire. Complete with destroying a new fast track career, taking my son, faking a murder, and creating a three ring circus of pain, chaos, and financial ruin (including chapter 13 bankruptcy). This lead to a second time of living on the bottom and having to start all over without degree, and no longer having a connection to a past that when referenced, looked bad despite innocence. I avoided contact, which was hard with a son, and who i would only see once a year maybe (rather than see the world or ever have a vacation), and the circus continued until she really stepped into the pile by attempting to rob a bank with three kids in tow, one of them mine. After that, things smoothed out, as she did, and my parents had the boy, but i had to pay double child support, or else he would go back so she can get her checks. I as a father had no rights, regardless of what she did or didn't do and i as a father, was considered ultimately responsible. Chapter thirteen bankruptcy, and such were natural occurrence as in not being able to build wealth.
Do not cry for me, or feel bad, as the song goes, life aint easy (and aint isn't a word). Somehow i managed to get an apartment in NYC, and did all kinds of interesting things that people say i should put in a book (ghost writer wanted.. i said ghost writer, not rider). Over those years i invented new tech, but my choice in partners has dogged me, and so, nothing came of it. it still sits there, including a way to search big data and process genetic information faster than anything they will do with Altera or Xilinx (mathematically proven), using off the shelf tech too. It can go through ten 3.5 gig genomes in a few seconds when current tech still takes hours. but i cant sell gold to a banker at discount. I did consulting work, and eventually took a permanent position which lasted 14 years 11 months, where recently they decided to dissolve my place, and put me out as my existence lowers diversity in the world.
Over these last years, a decade and more, i met a wonderful woman and took a chance and married her. She hails from Indonesia, her family is great, and whats left of my family, thinks she is too. We have our issues as all do, but i do not regret being brave enough to get back in the pool after crazy on a cracker. i DO regret not having children with her for she deserved better, and i will always be dead inside for that mistake in a long line of them. Over the years in NYC i not only worked in medical research, and education, as a full stack one man band, but i even became a celebrity photographer, and in a transitory way floated in and out of the arts as i did as a child or young lad. You probably have seen my work, i never got paid for, but twas better than destroying my kidneys hanging out in the bar just to have people around.
Of course there is so much more, like doing the macarena with Ruth Buzzy on the set of Sesame Street (and same night spontaneously doing comedy with Madeline Kahn by accident over a Swedish meatball), becoming friends with some great people, though distant they are. I still wish i could bring my tech to market even if i dont earn anything, the ability to sift single nucleotide polymorphism in seconds would be a boon, and i would give it to the world for free if they would not forget me. My son grew up to become a very fine man, currently serving as an officer in the Navy, and getting the degrees i couldn't get. I did work with geneticists and famous people, like Oliver Sachs, and was crushed by Schadt of Illumina sequencing fame (who needs orders of magnitude when one gets awards for incremental improvements?).
So here i am, with a basket of businesses that probably will never be, some really cool tech and chip designs, as well as other products that i hold till i see someone else make and market, then move to the reject pile, my career ended and finding out too old to program (despite a life of self learning i am told us old folk dont learn and adapt. yeah, going from dial phones to programming them, doing AI in python, learning how to write trading software, and staying at the top of the computer science game without help, shows how much that truth is a fail), and not diverse enough automatically (which my wife from another land finds confusing, sort of), am going to face being old and homeless again.
Here i am trying to reinvent myself in under a year, hoping to find contacts who may want to do the adventure of tech to market, and learning how to trade better in shorter time frames using my brain to disconnect from a business world that has decided to behave much like the soviets my father and family fled from so long ago, and perhaps finish the job of erasing my family by circumstance. Though now i have many children as family back in Indonesia i love and wish to help before i leave, and somehow hope to see a grandchild (but probably wont happen), and see my art work (yes, i am talented out the ears in so many ways, but it really means nothing, a let down for all those that believe that if they had it, things would be different in some great way), ideas, and such scratch the page of time.
There is and always will be so much more, and over time things will pop out and up, but the above is more than anyone needs to get to know me from a distance and then some. Over the course of this journal you will see me attack the problem of changing my world, and being the hero in my life's story that saves those i love, gets and keeps the girl of my dreams without losing her, and not minding that eventually, i have to go. Just remember, no one ever gets out alive! Raise a glass to not being erased, and lets go onwards to the future!
Lets hope my journal is a bit more interesting than most, and a publisher finds it and wants to do the story...
or that someone notices my ideas and says, there is someone that together, there could be a complete something.
But hope is not a good strategy, God makes his own way as others do, and only my action will achieve.
Sounds great eh? just dont catch me in a bad mood... i can make one cry worse than Jim Croce can..
Given i am now unemployed, I couldn't take the day off..
Installed Anaconda python on my system
Installed Visual studio as you need it to do python for CUDA
Installed the Pydev plugin on eclipse (to edit python, as i dont like visual studio editor for common things)
Put in the CUDA toolkit which you need to run CUDA Python..
Then had to add various packages to the virtual development version i set up
scipy, numpy, pandas, and on and on... including NUMBA which can compile python for GPU usage
busy busy busy...
The whole point was quant finance, learning Python to increase job offers, and CUDA machine learning
maybe if i do good enough, i wont have to go back to work... old fart that i am.
which brings me to the subject of the day...
understanding trading math...
To be better than a good trader, one has or should understand the game
and for me, understanding things like this viscerally is to understand the math and what it means
given you have approximately 250 trading days a year, give or take...
This depending on whether you trade Forex, Futures, or just plain old Stocks..
IF one wishes to day trade, depending on methods used, what are the odds?
what percentage of trades have to go your way? what can you get away with?
what are the most important variables?
Basically, sans anything else, at the time periods people day trade at, stock movement is close to random
regardless of that, its odds are 50/50... it can go up, or it can go down... it doesn't stay fixed...
so mostly its behavior is akin to flipping a coin mathematically. Though this coins odds do shift a bit.
How one looks at something to understand it is important...
you pick a stock, you pick an entry, there are two possible ways it can go.
and this leads to the question of how does one take advantage of such a system?
MONEY MANAGEMENT is the first lesson..
When someone joins or writes here that they blew out their account,
you know money management was not something they understood...
IF we took out a coin, and flipped it, it has 50/50 odds of its direction of play, heads or tails.
under normal circumstances the rarity of it landing and balancing on edge is not a consideration
even with stock, which could stay the same for a tick, its just a delay till a up or down is registered.
IF one can't change the odds of the game much, then how does one work such a system to come out ahead?
MONEY MANAGEMENT
what does this mean? well, if you say heads i get 10cents, and tails you get 10cents..
over time, the longer you play the more likely the payouts will converge to where you started..
you cant change the odds of the toss, but you CAN change the payouts!!!!!
Now if you say heads i get 15cents and tails you get 5 cents... the outcomes change...
even though the odds dont...
with 10,000 tosses (by computer since i program) with heads being + 10 cents and tails being -10 cents
the outcomes swing all around...
here are the results of 10 rounds of 10,000 tosses at that amount
1) -7.20
2) -1.80
3) 8.20
4) 2.80
5) 3.20
6) -10.80
7) .20
8) -3.00
9) -9.20
10) 15.40
quite uneven... however, they dont get very far from 0...
it also shows that 50/50 may be the odds, but in the real world monte carlo style
these odds dont come out HTHTHTHTHTHTHTHTHTHT
but more like HHTTTTTHHHTHTHTHHHTTTTTTTHHHHTHTHTH
another thing about those series of numbers is the purse size..
if this started with a purse of $10 instead of just adding up the outcomes from an infinite purse
Thats another thing that changes the game, as does the bet size..
with a purse that is large enough vs the betting size, the game can go on for a very long time
possibly forever... or so long the person tossing the coins would get old and die..
what if the the purse was 100 with heads being 10 and tails being 10?
how long would the playing go if the game stops when the purse ran out?
the question now is how many turns before there was nothing left in the purse?
here are the number of tosses before the subject goes broke..
1) 162
2) 48
3) 18
4) 52
5) 16
6) 96
7) 220
8) 162
9) 208
10) 62
no where near 10,000 flips..
what if the purse was 1000, heads was 10, and tails was 10, and the number of tosses was limited to 100,000?
1) 8976
2) 100000 1140
3) 6824
4) 100000 2260
5) 100000 2120
6) 56340
7) 6504
8) 32290
9) 100000 2820
10) 9022
out of the 10 rounds, only 4 of them lasted the whole way to 100k turns.
the rest of them crapped out at some point... so with a small purse, they all crap out quickly
with a larger purse, they take longer, but most crap out, and some run to the end
thats a lot like day trading results in some ways...
but what if we add asymmetry of outcome? which is in essence what money management does
money management says basically, if you lose, you lose less than you gain if you win...
now the interesting thing is that if one was trading options, the money game is losing about 4x what you win
but your odds are better than 50/50... while with day trading, your odds are 50/50 - skewed as best you can by trends, and waiting for entry points, etc... but nothing like with trading options...
(and kids ask what is math good for later in life. HA!)
so lets change this one thing... we still have a potential of 100k tosses limit..
but now, when you lose you only lose 5 , and when you win, you win 10..
we dont change the odds, but we DO change how the money is managed
lets first do it with a purse of 100... as before.
thats a bit different isnt it? even with a small purse of 100, look what the asymmetry did..
it didnt change the odds, it changed the outcome.
because now, the outcome in effect changes the effective odds..
next we will change the purse size to 1000...
what this is akin to is whether your under-capitalized or not
that is, does one have enough in the purse to make it through the bad runs to win the good ones?
before i show the results... lets recap the variables here..
The odds of the outcome are near 50/50 in terms of play, and in this case in terms of hit rate (we all know some traders do better than 50/50 in trades... however, given the right variables, 40/60 could do ok... )
the other variable is the size of the purse, and last, the asymetry of the pay out..
another, which we have yet to explore is the size of the bets themselves in relation to the asymetry..
this all would be akin to whether your account has 5000 in it or 25,000, or 100,000.
whether your stop is 10 cents down, and your target is 20 cents higher...
and of course... how many trades... of which 100k would not be possible..
not much change, is it? once the purse is large enough that one doesn't lose by running out
the amount at the outcome is completely driven by the MONEY MANAGEMENT combined with the odds.
having a huge purse, but betting the same, gets pretty much the same outcome...
if i made the purse even larger, it wouldn't matter... if the bet size was the same...
THIS IS HOW I UNDERSTAND PROBLEMS I ATTACK OR GO INTO WHEN I HAVE THE LUXURY TO DO SO.
with good money management one could throw darts at the board and still come out good..
this is why this is one of THE most important things... it even matters more than picking your plays!
for fun, i will run it one more time... but this time... it will be a 10 point stop, and a 25 point target
the first run will be a purse of 100...
100 point purse / 10 point loss / 25 point win
1) 100000 749435
2) 100000 752655
3) 100000 750450
4) 100000 749540
5) 100000 745165
6) 100000 747720
7) 100000 748560
8) 100000 745690
9) 100000 746040
10) 100000 746075
what if its a 50 point loss and a 100 point win? with a 100 point purse?
1) 2 0
2) 2 0
3) 100000 2498600
4) 2 0
5) 2 0
6) 100000 2499350
7) 100000 2486300
8) 100000 2472050
9) 5 0
10) 100000 2460950
we are back to some winning spectacularly, and others crapping out..
their bets were too large for them to last through the bad runs...
the ones that didnt, built purses large enough to last and so kept going to the end..
again with a 1000 point purse..
there is much more to this, and my next post will expand on this..
but this should give a person who wants to do this quite a bit to think of..
that their purse and size of entry into a trade have to be right to last through the bad trades
and if the size of the trades are too small, the outcome is limited by that in the long run, not the purse size
and i always love comments... just dont be insulting... i dont like to be vicious in return
you dont have to be positive.. or complimentary... etc... but dont be nasty, stick to the actual point
Well not quite a saddle, but this time, a double diagonal. I was looking at trades for MA (Mastercard) and SIG (Signet)… the latter of which I traded, and should have held since it gapped up a big amount.
Anyway… I was looking at some complicated options strategies… in this case, a double diagonal… the software gave the losses fixed at about $170, with potential gains over 3000…
However, I am now trying to figure out how to get out of the trade safely… while its pretty easy to do a trade that sets all the elements in place..
That is
Buy a strike 350 call exp Jan31
Buy a strike 350 put exp Jan31
Sell a strike 350 call exp Jan 24
Sell a strike 350 put exp Jan 24
The net price when I started was 37, but given the market I got it at .66 X 3 which is $198
This sets my maximum loss to $198 – the price I paid for the 3 orders
The point is to sell before the end of the week, and collect what I can..
Stay tuned and see whether our illustrious hero did himself in, or not… 😉
I decided that this call strategy was way too complicated for me given my being used to much less things..
Soooooooooo i took the 57 dollars and ran...
yeah, not much, but better to take that, then risk more...
a Double Diagonal is complicated as it works in time..
meaning that one has to let the first early set expire, and then run the second set
this complicates the calculations greatly compared to other kinds of plays...
tomorrow is another day...
now back to python, AI, Quant Finance... and maybe the next post on money management vs the rest.
as a hint towards that...
the play odds (50/50) dont exist alone once you add money into the mix and asymmetry
which was the point of the long post... that these things have to work in synergy not in isolation
without that understanding, one can win in ones mind, and seemingly on the table, but lose in reality
I am trying to find a good formula for computing odds with cash..
After all, in day trading and brackets, one gets to set the odds vs outcome
the point of the formula is to compute the minimum purse to succeed given a symmetrical play and an asymmetrical outcome...
When last we saw our hero, he was boring people to death who wanted day trading and short-term trading to be this sexy thing in which they grabbed the bulls by the horns, and rode off into the sunset of great wealth (while having fun too).
Pop goes that balloon… at least for the person doing it themselves. This is often the case in life, the doer is doing and relies on many things to make their magic, while the watcher sees the front they want to see. This is why people think talent makes things easy, not harder, and that talented people do the things they do because they have it easier and so, they can. This is far from the truth, as I have known many talented people besides myself, and every one of the really really talented, had a common quality of being hard on themselves (higher quality), working very hard (industrious), willing to give up on somethings to do other things (sacrifice), did not give up (perseverant), and would work to completion (follow through). Their talent was just a cover for these other things that people often ignored for whatever reason in pop psy or real psychology anyone wanted to bring up.
Well, some things that people are talented in, you really needed talent on top of those other things, and in other things, talent could be learned. Most people could learn to be very good musicians, professionals, and do very well, but few could be like Mozart, or the beach boys (oh yeah, don’t think they are in that list? Then you come up with a new sound that is unique like Scott Joplin did creating Rag).
Well, there are probably talented day traders, like talented video game players. However, most of them will admit that like the musicians, they work very hard at it, and gain an understanding of things. putting the hucksters to the side, who offer to make your dreams come true if you take their class, understanding is a critical part to things unless you’re a savant of some sort, and if you were, you would know it already if you bumped up against what it was you were savant in.
So, now where did I leave off? Oh yeah.. money management, the dead horse that those in the know keep beating, but do not give a visceral reason why. they convey a necessity without the reasoning or proof behind it.
We started simple… that stock has a 50/50 outcome like a coin toss… which it does, though its true odds may not come out 50/50, but may come out more like a loaded dice, or rather a loaded dice that can change its loads. At least once you figure out the load in the dice, you can win too, these constantly switch dice, and change loads and still, the outcome is only 50/50 in terms of heads vs tails, up vs down.
We should also have learned from the prior point that it is not whether this game has this kind of outcome that matters, or if the actual heads vs tails is not in true balance that matters, but that how the plays can be made that controls the outcome as much as the straight odds seem to. This is important because if you haven’t noticed, most of the hucksters out there are all about changing your odds and your picking, not changing the boring part of how you handle the cash purse!!! Exciting vs boring like above.
Heads and tails, as long as the bets were symmetrical, no one with even a huge purse could make it very far consistently, their fate was always with the gods. The problem here is that without consistency, one can’t earn well enough to make a good life, let alone reach some higher goals.
Combine the event, which one has limited control over, with the rules of the outcome, and everything changes. If losses are not the same size as wins, then outcomes do not follow the same rules. You can’t change the heads or tails toss, or peer well into its odds, but you CAN change how you create a win loss ratio. This has such an overwhelming effect that the even the size of the purse has no effect as long as the win loss amounts stay the same!! Once the purse size was above some minimum, the game could go on for 100,000 tosses, accruing over time! So the purse size only limits the rate from the start of how large the purse gets over time. Start with a large enough purse for the size of the bet ratio, and you will always end up ok. Start with a purse too small or a bet ratio too big, and you risk crapping out before the purse grows into the size that defines always ending up ok!
If one could figure out what the magical size was, one could always define either a purse size for a win/loss ratio or a win/loss ratio for a purse size. If one could find a formula for such, then increase the win/loss amounts in the game by half of that, one could, over time, sail to the point one wishes to.
That’s my next stop… taking what was in the post on money management and start to figure out the math
Boy are my numbers going to be bad today... ha ha...
Paper trading and trying too many things, and not being serious enough...
sometimes we all need a break..
As for swing traders i have decided to put up a list of stocks and track the outcome based on my new coded strategy.
of course no one is listening to this for the most part... so i am unsure who i am talking to...
I am testing a strategy.. so whether you decide to follow or not, is up to you... Caveat Emptor...
Date Company Ticker Shares Trade Price Cost Open/Close
1/24/2020 MPLX LP COM MPLX -1000 24.81 -24810 Open
1/24/2020 Covenant Transn GP com CVTI -1000 13.79 -13790 Open
1/24/2020 The A2 Milk company LTD ACOPY 1000 10.47 10470 Open
1/24/2020 Fortress Transportation FTAI 1000 19.88 19880 Open
1/24/2020 AVAYA holdings AVYA 1000 14.06 14060 Open
1/24/2020 Companhia De Saneamento SBS 1000 14.82 14820 Open
1/24/2020 Broadmark Realty BRMK 1000 12.73 12730 Open
3000 33360
So here i am...
writing code every day for various end reasons, and a singular (mostly) goal..
needing to earn a living so i can keep going..
given one investment may come in with a 70k profit this friday, i may have bought time..
one resource i have is 13 years of stock day data.. (open, high, low, close, splits, dividends)..
and this covers 11,000 tickers... i am lacking part of 2019 and of course 2020
now that most sources are locked up and i have no cash to pay...
those that are free are quite limited in their output to drive you to pay...
eventually i will... but do i really need YTD data to do machine AI learning?
not really... even less so that i have a titan x card, and 30 plus years in applications engineering and data
today i think i am going to see about data quality... meaning i will try to map out what i am missing (if anything)
and go about slowly acquiring it..
i am also going to check into CUDA python, NUMBA and Pytorch...
with a goal to have a test set created by later this day..
with 251(days) * 13 Years * 11,000 ticks of information
it would be a shame to waste 35,893,000 days of information..
Yes... the Title says it all...
when your going to do machine learning AI... deep learning... or even just cool statistics..
what you want is DATA DATA DATA..
I am lucky enough to have the foresight to have obtained years of data...
[i may even have more that i have yet to process]
going through what i have in the database already, i am working on 2007 (not 2016) to 2019
its going to be good enough...
but its going to take me a long long time to get this set up...
the database is the source, and i am creating a new DB for this project..
and the data is now being put into rows of a table with the precise goal of having a base to generate files for machine learning projects i have designed..
To give an idea... the current High Low Open Close data is 19,954,035 rows in a MSSQL DB
I am a lot of things in real life... not just a data guy..
but in data and programming i am a geeks geek...
[tons of languages, tons of databases, tons of real world professional experience,
all starting back at bronx science in the late 1970s... ]
I just ramped up my Python for finance... (doing quantiacs and quantopian on the side now)
and so, now i am getting into machine learning (AGAIN)
what do i mean by again? well i have played with it off and on since the mid 1980s
back then you had to cook your own systems...
so i let it rest until recently... why? because now its a mature thing and i can focus on projects
and not focus on all the stuff that has to be written and handled and so forth before you even think of projects!!!
And i have a lot of interesting projects i can now get into..
not to mention i have a machine that can handle this without having to rent cloud space..
Sadly i have a first generation Titan X...
[i will have to look into whether i can add a different Titan Series to the machine with it, and what i can do if i want to expand... but i can add three more to my current box... and will add at least one if i make enough cash either taking a new job, or online, or both... ]
well... have to go, more work on the data has to be completed
takes a long time to run it, and of course, finding an error, and you have to run it all over again
the more i get into the final table (not the test model table), the longer it takes to rebuild it if there is an error
This lead to a second time of living on the bottom and having to start all over without degree, and no longer having a connection to a past that when referenced, looked bad despite innocence. I avoided contact, which was hard with a son, and who i would only see once a year maybe (rather than see the world or ever have a vacation), and the circus continued until she really stepped into the pile by attempting to rob a bank with three kids in tow, one of them mine. After that, things smoothed out, as she did, and my parents had the boy, but i had to pay double child support, or else he would go back so she can get her checks. I as a father had no rights, regardless of what she did or didn't do and i as a father, was considered ultimately responsible. Chapter thirteen bankruptcy, and such were natural occurrence as in not being able to build wealth. 
