ITS CALLED STATISTICS:
There are thousands of cunt coins out there.
Their prices are normally distributed as per the central limit theorem.
Find the mean price of all coins, buy all coins below that price to make gains. They will steer towards the mean, eventually, as more shit coins enter the market. Easy cash.
> in b4 assuming independence.
you're welcome.
Other urls found in this thread:
coinmarketcap.com
twitter.com
>price
you mean marketcap, right?
Explain it more in detail, please
>Find the mean price of all coins, buy all coins below that price to make gains. They will steer towards the mean, eventually, as more shit coins enter the market. Easy cash.
You're making a lot of assumptions here and most of them are dumb.
they are indeed dumb. obviously, the coins aren't independent. any rise or fall by big brother btc leads to massive alt coin swings.
i am also assuming that new coins that enter the market start of low. this is not true, just look at CAN. $5 initial buys on ku klux coin.
This is bait r- right?
no
can confirm, this has been a very profitable strategy for years, although it is just a little more complicated than OP's description
yes and no.
you get bombarded with information constantly. news here, news there. a tweet flying at you, and you immediately want to buy.
but there are established rules that are inherent to every system, so at times i think it is appropriate to look at the global picture here.
another such law is self organized criticality. systems, ranging from as large as the galaxy to ocean currents are subject to perturbations across all frequency ranges. however, the power distribution tends to have a 1/f distribution, implying equal noise power across each octave. what does this mean? low frequency perturbations (moons,crashes) dissipate into higher frequency oscillations (+/- 10% changes). so here's an example - a moon in one coin will dissipate to high frequency garble in other coins as the system tries to reestablish equilibrium. finding the coins that mooned alongside the coin of interest, and finding the coins that didn't do much will be crucial for determining when to invest.
i might add, this is true by definition across all time scales
how do you crunch your data fellow voodoo magician
Returns are not normally distributed dumbfuck!!!!
i dont, need to start doing it tho but dont really have the time.
i am literally making all of this up based on some stuff ive read recently. i have 0 econ knowledge.
>coin prices are random variables
>coins are independent
>value approaches mean proportionally to time on market
Ouch!
That being said, OPs plan will work but for none of the reasons he listed. Anything cheap will inevitably moon sooner or later when the normies realize its their chance to buy Bitcoin 2.0 at $100.
See: TRX, XVG, XRP
but it is the bellcurve of reestablished equilibrium...
Same thing I do with stocks.. shh..don't tell wall street!!!!
y-yeah exactly bro
Of all the things I have seen on Veeky Forums, this may the dumbest. Please cash out before you lose all your monies OP.
if this is based on just recent findings, my man you'll make it in record time
but ill try give a first pass at what id do.
i work in comptuational neuroscience and one key objective is to find clusters of neurons which are connected to each other. i.e. if neuron A spikes, which other neurons are most likely to alongside neuron A, shortly after, and which neurons fired shortly before.
so i'd start by creating a high dimensional parameter space, with paramters being information that includes power spectral density, mean spike rate, connectivity metrics, etc. then do a pca decomposition to reduce complexity and find those clusters. but im just pulling this out of my ass.
what he means is buy low sell high
should i buy PAC
that shit is gonna stay at 1 sat for the next year.
that is absolutely wonderful
these types of eclectic research strategies really pay off in crypto thanks to all of it essentially being data
i'm not sure how many trades could be based off of this, but just seeing a visualization of that would -really- make the whole closed ecosystem thing click and make you really -feel- the market
this is wonderful, may your moons be plentiful
>ITS CALLED STATISTICS
nobody here understands statistics, if the previous nigger hate threads were any indication
hmm what features are u looking at here?
how many components when u do pca decomposition and how do u deal with the computational complexity of u scrape an entire db say cmc history.
lets say u reduce it down to 2dimensions, cluster it using some hierachicsl clustering or even kmeans and plot it so u can visualize it. how do u use this information to help u trade? how do u verify that these cluster of instances (in this case shitcoins) are the high profit ones? this seems to be an analysis of current and pst data which is different from prediction of future of which this deals with time series analysis.
This has nothing to do with CLT you dumb fuck
hey man i'm a MS-level biomedical engineer/biostatistician and my job is pretty much dealing with heaps of raw data from our research labs and writing custom code to chew it up. This includes data ranging from AFM measurements, genetic studies, genomics/proteomics, etc. I am willing to spend my off-time working with you to develop a program to indicate to us when/where to invest--i am not interested in selling our work, only using it to make gains.
if you want to go in together leave a burner email or something. i'm not ever going to scam you or ask for trading info up-front. i really want to apply my autism to make $
lmfao whitey think's he's better than a nigger
your average iq is 100 that's borderline retarded
hey i'm msee with focus in signal processing and ml. wanna collaborate?
throw out some ideas but the only thing that worries me is taxes on trading (don't fall for the not paying tax meme.. no point i. making it if irs rapes you in 20 years- this is possibility since crypto has gotten a ton of news and gubment wantstheir cut)
black iq mean is a full standard dev below white iq mean. if avg iq of 100 is borderline retarded then what is avg iq of 85 (nigger iq)? potato? try again, sweetie
>the only thing that worries me is taxes on trading
>use offshore exchanges with little verification
>make all trades through a VPN
>store long-term gains in xmr
no taxes
will def consider! [email protected]. what school you studying bme & biostats at?
i got to get to work so im gonna bail this thread. feel free to hit up that burner email above, would be happy to discuss more!
...
I was shilling pfr and prl here for you on /biz since both were 0.07 and 0.01 respectively. Get profit on egas and make me (and you) rich.
Take a look@ coinmarketcap.com
My tensorflow brotha up in here
also, all of cmc data is redundant in my opinion.
the current state of the market is much more profitable and ML can actually flourish in that.
These distributions are not Gaussian, they're Mandelbrotian
They are subject to 80/20 rule, the top 20% are going to make 80% of the gains and the rest are going to make 20% MAYBE before they fall to zero
>hurr durr we're smarter than literal retards that means we're smart!
stay borderline-retarded, yt