Technical analysis doesn't work

Technical analysis doesn't work.

Prove me wrong.

TA has been around since the 17th century and every investment management firm uses TA to some degree in their work

It is bullshit, but because people use it, it can come true. Self fulfilling prophecy type stuff.

it probably does under some crcumstances, but you wont be profiting off it because u dont have the resources or smarts to stay ahead of the technical analyisis metagame

Yep, doesn't work, ignore it goyim

Interesting

Price is a technical indicator.

P/E is a technical indicator.

Volume is a technical indicator.

Without technical indicators and analysis, you couldn't have valuation formulas and fundamental analysis.

What is exactly is technical analysis? Obviously you can identify trends using the charts, but a north-facing-eagle-holding-a-crab doesn't mean anything.

The history of the chart is physically impossible (at least within this Universe) to contain concrete information regarding the future. Always remember that while you're on your grand quest of discovering the next memegrail. You can fully predict the movement of a stock, up to the last cent, but the chart history isn't sufficient data, nor is any other data. You must have a full knowledge of the entire environment that affects it to do so, which means the entire market. Everything less than that is casino and carving an extra percent to your 50/50.

its bullshit in theory, it works in practice because everyone is basing their trades on the same principles. ie. traders are responsible for resulting price action not the markets

also,

clueless retard

>he doesn't understand how volume works with price action

Enjoy being eternally mapless in the markets

No
Fuck you

If something has been consistently rising in price for the last 3 days there is more than a 50% chance it will continue to rise.

what am i even reading
just because there is more than 50% chance that it will continue to rise does not mean you will make money off it in long term, because u can hold a stock that raises for 10years straight and fall for 1day and u are in the red

You're pretty intelligent and should keep posting as if you have advice to give.

>basing analysis on past 10 years
>looking to exit trade based on outcome of a single day

genius.jpg

my point was that basing your analysis on just 'rising' or 'falling' prices is not enough because raise in price often happens over long periods of time and falls over a short period. The example is obviously an exxageration

>lol I know how future price moves will shake out
>but you're an idiot for using technical analysis to determine future price moves

You're intelligent and should keep talking as if you have advice and not inane rambling to add.

>lol I know how future price moves will shake out
ive never said any such thing, i said stocks tend to fall faster than they raise, which is i think very uncontroversial

quality posts mate, thanks for contributing to the discussion

>he still doesn't know how volume affects price action

Lol

If I was using other peoples money, I would trade

With my own money, buy and hold

Technical analysis is just quant analysis without the statostics, modelling, simulation and backtesting.

Essentially, it is the astrology of finance.

TA works because everyone is doing TA.

Resistance/support levels are fucking amazing for trading and that is TA.

What's wrong with you, OP? Learning TA is an advantage and every advantage is needed when trading.

>Prove me wrong
Lol.
Why would I want to do that?

This is true, people trading in reacting to TA in sufficient numbers is what validates it. It's pretty much the real-life equivalent of the waaagh.
This is also how institutions make big gains, they know candlestick patterns attract a cluster of stops above/beneath them like flys to shit, they have deep enough pockets to push price just enough to trigger sufficient stops that perpetuate the move scaring off any momentum in the opposite direction ultimately leading to a reversal.

Its what separates profitable traders form losing traders, the market is random, people arent

>traders are responsible not markets
Youre retarded

stocks do all sorts of crazy shit in between quarterly earnings reports. It's amazing to watch.

It is also completely random.

ITT: Retards

Using minimal assumptions (that are true 99.99999% of the time) it can be shown that TA cannot be true using mathematics.

Spoken like a true gambling addict.
That's bullshit and you know it.

Some stocks do exhibit consistently high returns, but that increase in price is largely the result of exposure to systemic risk factors like market prices, macro risk, liquidity risk, etc. You can't get consistent returns without taking on risk.

>memeing yourself that it doesn't matter when you buy
>even though historically its proven that end of the year is best time since managers start to pad up their portfolio

>what is trend and bias can affect fundamental values

You humongous retard, you're mixing a technician with TA.
Quants use Technical Analysis, mixed with the behaviour of an asset in a statistical way, and use it to get advantage of the TA modelisation to generate profit by "minimum risk taking". Quants are more backtested, both use technical indicators in very similar ways.

You will never have the full data, read about portfolio theory before you spout more stupid shit.
Prove it, your claims have no hold unless you can prove them.

>He thinks the market, traded and by thousands of high powered computers, is completely random.

He thinks thousands of high powered computers using Technical Analysis are wrong.

topkeker

>you humongous retard

I'm a quant, dumbass

depends what you mean by TA - if you mean a very wide definition that includes any investment decisions made based on price data then sure... but the sort of stuff used by HFT prop firms and systematic hedge funds is a far cry from the chartist stuff used by retail traders

P/E isn't a technical indicator - the 'E' part is going beyond technical analysis, that is fundamental data

Thank you for the first sensical post of the thread. I appreciate it. Riskier stocks are by all nature higher profit, but also have a higher variance and it should be obvious to anyone who understands anything about asset pricing.
Does this mean buying solely by TA is smart? No. Unless you're building a diversified portfolio (before any of you mouthbreathers come complain about the evils of portfolio diversification, fuck off we're not talking about your 401k but an actual fund that needs to allocate so their risk is acceptable to their investors. Most positions held by big funds are positions they don't actually want, they just need to please stupid retards like you, who are sure they know how to invest after reading the summary of Intelligent Investor 3 months ago.)

But TA is a way too large of a field to totally condemn, it's ignorant to do so. It contains some aged principles that are indeed the equivalent of Astrology, but as I outlined above, it's something that's expanded upon in ways that work. How would the success of Two Sigma, PDT Partners, D.E. Shaw or Renaissance Technologies , let alone Marty Schwartz, Mark D. Cook or Jason Perl exist if TA would have no working models or principles? All of these mentioned have out-performed consistently, with obvious dips, the stock market. Does this mean there are no retards who swing around TA models and bullshit to generate retarded investments? No. It just means retards exist and don't know what they're doing no matter what they're doing.

>Learning TA is an advantage and every advantage is needed when trading.

nope, any 'advantage' or edge will do...

You use TA and you don't even realise it? I must have misunderstood you, are you implying TA is a less sophisticated version with missing bells and whistles, if so, yes you are correct.

I would anyways be very glad to hear a quants side on all of this, how do you feel about funds that have started to expand on A.I trading? Are neural networks accepted yet as a valid idea to work with?

>Are neural networks accepted yet as a valid idea to work with?

neural nets have been tried (and often failed) since the 1980s

RNNs show some promise on price data, there is also some research in deep learning and alternative data sources - though that, be definition, is outside the scope of TA.. Machine readable news started generating some interest too about a decade or so ago, again outside the scope of TA.

Of course those are out of the scope of TA and that's why I was asking about the neural networks, Machine Learning has had huge leaps in the recent years. Especially a lot of buzz with generative adversarial nets, (though I doubt there could be a lot of use in trading with that).

It's not like neural networks are a static thing, there are multiple architectures and their functionality depend on training algorithms.
Do you see any future for them helping to understand what makes markets move ? ( Considering this is the core problem of all securities trade)

well those are all examples of where neural nets have actually been useful... it just isn't in areas related to TA

their functionality depends on their structure not whatever algo (mini batch SGD, Adam etc..) you use for training - that simply governs how long it takes to train them and how well you can avoid local minima

I'm not sure they're going to, by themselves, provide further insight into what moves markets

This is a very interesting statement you've made as a whole, considering the 2016 quant summit europe was mostly centered around the usage of A.I in trading and 2017 will have quite a bit about machine learning.

Their architecture does indeed assign them new functions. So the implementation of algorithm and usage of data varies, it affects a lot of things according to all of my understanding, the accuracy gotten does have an effect on data quality (training cycles), and how we choose to train it affects does it lead to correct predictions .

Thank you for your time, you hold a very interesting opinion which might need a bit of updating.

sorry if this appears rude but is English your first language and have you ever build a neural net?

You've made some statements regarding training neural nets and it isn't at all clear what you're talking about.

>have you ever build a neural net?
and now, ironically, I'm making myself hard to understand...

*have you ever built a neural net?