Technical analysis

[Harold O'brien]

@Prince88

My gripe with technical analysis is the same as conspiracy theories and alternative medicine.

If alternative medicine actually works it is called medicine. If conspiracy theories are plausible they are called theories. If technical analysis makes decent predictions it is called financial modelling (i.e. genuine mathematical models with reasonable assumptions, not trend lines or candlestick patterns.)

 

[EverythingMattersCel]

"Technical analysis" with respect to retail traders manually drawing a few squiggly lines on a chart is the astrology of the trading world. Maybe it's a self-fulfilling prophecy in nascent markets where there are only naive, retail fish, but in traditional markets u are quickly eaten by sharks (institutions with superior trading infrastructure and knowledge).

Technical analysis in the sense of simulating 1000s of indicators to implement an automated strategy with algorithms is legit. That's the kind of "technical analysis" institutions do to build their models.

 

[Deleted member 19896]

wat

 

[Crusile]

If alternative medicine actually works it is called medicine. If conspiracy theories are plausible they are called theories. If technical analysis makes decent predictions it is called financial modelling (i.e. genuine mathematical models with reasonable assumptions, not trend lines or candlestick patterns.)

Most conspiracy theories are low probability but impossible to be 100% conclusive of. The difference is you can validate it with 100% authenticity yourself, by learning it, then deciding. Your conspiracy theory is that there's tens of thousands of people lying about ta and making fake screenshots on profits idk. obviously this is a bigger conspiracy theory.

but agree long term price predictions are useless. its just binary 50/50 probability and outcome because high variance high variables

 

[Harold O'brien]

Most conspiracy theories are low probability but impossible to be 100% conclusive of. The difference is you can validate it with 100% authenticity yourself, by learning it, then deciding. Your conspiracy theory is that there's tens of thousands of people lying about ta and making fake screenshots on profits idk. obviously this is a bigger conspiracy theory.

it's impossible to be 100% conclusive of anything

there are also tens of thousands of people with losses that keep quiet about it

 

[Crusile]

it's impossible to be 100% conclusive of anything

there are also tens of thousands of people with losses that keep quiet about it

What if you try learning then become millionaire/billionaire with ta only.
yes its a zero sum game, you have to preform to make money.

 

[brucel]

are you actually a leftist? you talk like a little bitch

 

[Harold O'brien]

What if you try learning then become millionaire/billionaire with ta only.
yes its a zero sum game, you have to preform to make money.

no it's not - this is what finance noobs say. at the fundamental level resources are being generated

 

[Harold O'brien]

are you actually a leftist? you talk like a little bitch

intimidated by my intellect, brutal

 

[hebbewem]

are you actually a leftist? you talk like a little bitch

He is

 

[Incellectually_Shy]

wat

This can be read in seconds.

 

[garfyld]

Technical analysis is form of apophenia

 

[Crusile]

no it's not - this is what finance noobs say. at the fundamental level resources are being generated

It is, ta traders trade small price fluctuations. you need to outcompete other small traders, algos, funds. Non ta traders add to the prize pool but its still competing vs other traders for it- close enough to zero sum

 

[Harold O'brien]

It is, ta traders trade small price fluctuations. you need to outcompete other small traders, algos, funds. Non ta traders add to the prize pool but its still competing vs other traders for it- close enough to zero sum

aka completely random bullshit

lol at muh skill

 

[Unhuman]

It's the thing you do when you're too lazy to do fundamental analysis.

 

[Crusile]

aka completely random bullshit

lol at muh skill

it takes 2 to tango, you need to learn it to disagree with it

 

[vox2992]

If you guys want to get a head-start in the quant landscape, go into Pure Maths and learn about Topology (University level calculus -> Abstract Maths -> Real Analysis -> Algebraic Topology, is a good start). Modelling data spatially will complement current Deep Learning advances. From newbie level it will take about 5 years of training, but if you have the ability to apply said principles to real problems, you're guaranteed a certain level of success.

 

[Harold O'brien]

If you guys want to get a head-start in the quant landscape, go into Pure Maths and learn about Topology (University level calculus -> Abstract Maths -> Real Analysis -> Algebraic Topology, is a good start). Modelling data spatially will complement current Deep Learning advances. From newbie level it will take about 5 years of training, but if you have the ability to apply said principles to real problems, you're guaranteed a certain level of success.

why the fuck would you do that

focus on ito calculus and computers

 

[Harold O'brien]

I can;t think of anything more cringe than being a quant though

 

[Harold O'brien]

it takes 2 to tango, you need to learn it to disagree with it

"you can't disagree with the christian god unless you've read the entire bible"

 

[vox2992]

why the fuck would you do that

focus on ito calculus and computers

Guaranteed multiple 6-figure salary if you're picked up by the right house and have the pedigree. The most famous publicly known one has a net worth of over $20bn. God knows how much they're making on the side.

Topological Data Analysis is still in its neonatal stage, meaning appropriate entrants bear a first-mover advantage. And we all know that to be successful in this world, you either need to be first, be smarter, or cheat.

Not saying it's the be-all-end-all, but it's a legit trajectory you can take if you have the talent.

 

[Harold O'brien]

Guaranteed multiple 6-figure salary if you're picked up by the right house and have the pedigree. The most famous publicly known one has a net worth of over $20bn. God knows how much they're making on the side.

Topological Data Analysis is still in its neonatal stage, meaning appropriate entrants bear a first-mover advantage. And we all know that to be successful in this world, you either need to be first, be smarter, or cheat.

Not saying it's the be-all-end-all, but it's a legit trajectory you can take if you have the talent.

Seems complete overkill for a field that is glorified guesswork.

Besides has there been any studies comparing the efficacy of these models to the standard stochastic volatility ones? If not this is an extremely retarded thing to do. Most likely you will spend years learning very complex math you probably don't understand for absolutely no reason

 

[vox2992]

Seems complete overkill for a field that is glorified guesswork.

Besides has there been any studies comparing the efficacy of these models to the standard stochastic volatility ones? If not this is an extremely retarded thing to do. Most likely you will spend years learning very complex math you probably don't understand for absolutely no reason

By definition if you're using a new method there won't be empirics to back it up. If there were, then everyone would be in on it and you'd lose the advantage. The point is rational properties that can be applied to appropriately modelled real-world problems to generate novel returns.

 

[Harold O'brien]

By definition if you're using a new method there won't be empirics to back it up. If there were, then everyone would be in on it and you'd lose the advantage. The point is rational properties that can be applied to appropriately modelled real-world problems to generate novel returns.

depends on how you define new

either way most new stuff never goes anywhere so it's likely a waste of time. learn topology because it's fun/interesting, not for an autistic gamble (I don't see it paying out) on becoming the #1 math slave of big hedge fund

 

[vox2992]

depends on how you define new

either way most new stuff never goes anywhere so it's likely a waste of time. learn topology because it's fun/interesting, not for an autistic gamble (I don't see it paying out) on becoming the #1 math slave of big hedge fund

Forget about hedge funds. In general, whenever we think about data, we think about X-Y cartesian planes right? Have you ever asked yourself why that is enough? If you've ever learnt the Numpy library, you'll know that you can play around with multidimensional arrays. Have you ever thought about modelling data spatially?

 

[Harold O'brien]

Forget about hedge funds. In general, whenever we think about data, we think about X-Y cartesian planes right? Have you ever asked yourself why that is enough? If you've ever learnt the Numpy library, you'll know that you can play around with multidimensional arrays. Have you ever thought about modelling data spatially?

I am aware spacial dimensions above 2 exist yeah