Can science predict the stock market? No I don't mean using the bullshit that economics teaches

Can science predict the stock market? No I don't mean using the bullshit that economics teaches

Other urls found in this thread:

youtube.com/watch?v=hV9rFwwAMxI
en.wikipedia.org/wiki/Black–Scholes_model
twitter.com/SFWRedditGifs

pic related is the performance of Renaissance Technologies, a hedge fund that almost exclusively hires only PhD mathematicians and physicists

you mean like high-frequency algorithm trading?

yes, check numer.ai

economics is inherently based on assumption, so no.

>economics teaches
>predicting the stock market

It literally teaches the opposite.
Kill yourself, brainlet.

there obviously isn't a way because otherwise everyone would be doing it

Yeah, you use factors outside of basic economics that influences it such as news and dialogue among the public. If you could get a wiretap on prolific members of the market and run machine learning algorithms which evaluates dialogue (kind of like how we have ML methods which evaluate and generate news articles) and combined that with current methods to predict stock trends, you could probably write an automated trading algorithm which gives you incredible foresight and ability to game the market. Like if you worked at Google for instance and had access to the business Gmail accounts of big players in the market, you could read bad behavior by management that you wouldn't have gotten a read with playing the stock market by looking at the numbers alone.

Deterministically, no, but there is some sort of prediction when it comes to the stock market.
Doing it will not give you any advantage, because it is what people who invest usually do. In fact, not doing it will put you at a disadvantage.

yes what do u think quantitative analysts do. it's not a perfect science
youtube.com/watch?v=hV9rFwwAMxI
en.wikipedia.org/wiki/Black–Scholes_model

>there obviously isn't a way because otherwise everyone would be doing it
only correct answer.

>they're not doing it yet

>doubling your investment every few years

jesus christ that's a hell of a fund

The problem is that a 'perfect' system of predicting the stock market would tell everyone to buy at the same time and sell at the same time thus negating the perfect system.

A system which predicts more wins than it does losses probably exists, but it's never going to be shared with the general public because it'd defeat the purpose of having such a system. You only win when someone else loses.

well renaissance is THE best hedge fund around. though they are closed to outside investors now so only their employees reap the gains.

> Economics teaches only random walk!
kys pleb

> value investing
> technical analysis
> factor analysis
> etc etc etc

>A system which predicts more wins than it does losses probably exists

A high ratio of profitable to unprofitable positions does not necessarily mean a system is good. I could buy hundreds of way out of the money options for a few cents a piece. If only one of them is in the money by the time it expires, it would likely more than make up for the hundreds of worthless options I purchased. Nassim Taleb made a fortune in 1987 this way.

Conversely, the vast majority of Long Term Capital Management's trades were profitable, however the profit per trade was generally extremely small. It only took one catastrophic loss to destroy the whole fund.

Proper position sizing and sound risk management are by far the most important aspects of trading.

What area of math should one get into in order to work in finance or (quantitative analysis?)

The most obvious one is statistics, but really any quantitative field will work. Many funds hire electrical engineers because of their expertise in digital signal processing. Even if you decide to study the more abstract and theoretical areas of math you could probably make it work. Ayasdi is a data science firm that uses topology to analyze huge data sets, and they do work for many financial institutions.

Thanks for your candid response user