>>16555525
Every automated trading bot on the market now (most of which make a ton of money) use probability theory and stochastic processes to help them make purchase decisions. Yes, learning the basics of probability, statistics, stochastic processes and time-series analysis will be helpful. There's no guarantees in anything, but they will be on the whole useful for analyzing market trends and making useful short-term price predictions (assuming no major disruptive distribution changes).

>>16555535
>probability, statistics, stochastic processes and time-series analysis
this sounds way to convoluted for the average investor to work, but you could also reply that the average investor is an idiot and it's there to get tricked.
could you explain how this stuff would be helpful?

>>16555525
Statistics with no causal modeling are risky, but they can help. You 'just' need to identify robust trends and be smart on how to bet on them.

>>16555525
Stats , and advanced stats are good for seeing where to look (but most of these positions are obvious low-hanging fruit).

Stats are good for seeing competence when you come across a stock and check the rivals.

Stats are good for entertainment/ fans

>>16556399
what's good for matket then?

>>16556400
If 'matket' I don't understand.
If market you have to see and understand the major models that drive the market. Then identify a few models, wait find a good one in your Competence, be disciplined, often then wait, if you found value Decisively capitalize.

bumping once, if this doesn't get any discussion I'll let it die.

>>16556395
The question of "when is it helpful" ultimately comes down to the same situation you have with people asking when they'll use algebra/calculus in their daily life. In most cases, you can get away with not using it, or using the absolute basics.

However, I'll give a few simple examples of ways in which these can be used.
Probability distributions give a way of modeling the changes between stock prices. In a very unsophisticated way, you could design a system which takes the average of the changes between price over a significant enough period of time. If that average is positive, the value is growing over time. If that average is negative, your value is decreasing over time. (no duh). Generally you won't see large enough variations in stocks that the median is required, but you could also use medians if your data is really all over the place.

You could then look at how a window of sample averages is changing over time. The plot of how your 30 day average changes over time gives you a way of seeing how volatile a particular investment choice is.

Sample variance/sample standard error gives you a different way of quantifying how volatile the particular stock changes are (in the short term).

These approaches, however, all assume that the changes in your price are independent from each other (which generally they aren't). If you wanted something more sophisticated, you could use an ARMA model to allow for "inertia" (changes in the past influence changes in the future).

None of these approaches are required. All can be helpful. These examples are looking at only a single stock over time. Things can get even more complicated if you're comparing multiple choices for investing the same amount.

>>16560233
wouldn't this get us in the "data jugglin" scenario? just messing around with loads of data in hopes to get and edge and/or biasing yourself because your stats told you so?
I think quantitative investing should be something but the more I look at market game the more it looks like rng and noise.

>>16560297
I think this argument doesn't hold a lot of water. Yes, if you run statistics, you could be misled by certain models (especially if you don't understand how they work and what they assume). However, if you don't take any quantitative assessment at all, you are essentially doing the same thing as picking a completely arbitrary decision.

Statistics aren't everything, but the "market fundamentals" are based on statistics. All of the "trend analysis" etc. that people use for qualitative judgment, are in some level approximations of statistical assessments of volatility and long-term time-series patterns.

Think of it this way. A doctor who has gone through medical school could be misled by an incorrect study into believing a particular treatment works when it doesn't. This is a risk. It's still probably less risky than having someone that doesn't have medical training or experience treating you.

>>16560297
>I think this argument doesn't hold a lot of water.
just to be clear, I talk with zero background on investment and what probably would be considered basic statistic on this board, that's why I rather read than post and learn something, because I have no clue.

>>16560389
>Replying to myself quoting another anon
I didn't have to be this retarded but I went all in