Different input: multifactor vs single (or limited) factor, and different processing of the (already different) input. I have a different opinion and experience with this than you. A computer will beat a human in a chess game, or in similar stuff with a limited set of fixed rules. But in a complex arena with thousands of different factors, all interacting with each other... I believe 100% that a computer cant match the human brain yet and will not for a very long time. A very simple example: to pet somebody, caress with the hand, so it makes the other feel really good. No computer, no machine, no robot can do this the same way as a human hand. There are millions of neurons, motorunits, hormones, micronuances involved, all interconnected with each other. No computer can process stuff like this like the human brain. Like I said, thats my belief and experience. If you have a different opinion, I accept this. We all have a different perception of reality. Basically, we all form our own reality. You see down, I see up. Thats great. Thats what makes a market.
so posts require documentation now? and your family livelihood depends on it? my hands are shaking now. the industry stats are out there, 95% lose, 1% make enough to live on.. plus the fact that it gets more difficult when the frequency gets higher - slippage, commission... satisfied now? by the way, by the same stats, roughly 95% of the stuff on ET, or the internet for that matter, is BS... what are you gonna do... documentation aint gonna help. actually the most valuable stuff on ET or the internet, is the BS... that stuff really trains your sense of smell, pun intended, so that you can sift thru and find the truth. this game is all about intelligence and counter intelligence and counter counter intelligence.. I show you smoke, you show me mirror... that's what the pros do... if you smell enough BS, you will find the truth, with some independent thinking. I have posted else where about the BS stories they pushed so far this year, and I was able to take advantage and pick stuff for cheap.
we are way beyond chess already... AlphaZero figured out the truth of chess, with 4 hours of self learning from scratch... it figured out the most complex game of Go... the google voice AI is indistinguishable from the other end. yeah in the future the 'petting' job may just be the only thing humans can do, only because the recipient knows it's another human.... if done remotely, you won't be able to tell the difference. these DOM stuff... it's really no big deal.. DeepMind just has bigger fish to fry at the moment.. and the Wall street guys are probably already doing it... perhaps some human traders are still making money... I know somebody trading like this too... but in my view the days are numbered... human brains will be no match.
price paints the chart realtime………..learn to read the promses made by the arrangements of the bar tips..….they reveal what is next super high percentage of the time.
Thanks CALLumbus, In my opinion, and my experience trying to program an algo: The problem is time and the many times of clicking backtest button. Every little change to increase the performance results leads to false predictive behavior of the algo. Another thing with algo is the time loss from starring at charts. I would spend hours programming and no time watching the live markets. I eventually return to mechanically intraday trading with the thought process and motto of "Everyday, lets make some money with my own brain, trade what I see and what I personal learn from and what I seen the past" In my opinion, algo trading was taking to long to get to the money. I agree. And what I learned is that preparation and controlling risk is very good. With one contract, I risk between $200-$300 per trade. On average its $200. The way I see it is, I see what the algos is doing by looking at the chart in real time everyday. I honestly do not care about the algos. They dont't pay me. In my opinon and my personal experience, most people fail at trading because they trade all their real money away before learning to trade.
Thanks CALLumbus My style is a bit of everything: systematic and experience, but most importantly I like to "get aboard the trend of the moment and choose my stop loss and profit target" I just want to take the trade and management it. That's my job. My job is to analyze the situation, wait for the price to move in a direction, make some money when it moves. It's really just that simple. When i am in a trade, managing it is the upmost importance. I use my predictive price analysis to guess where I think price will stop and price may go the other way, then I exit for a profit. If I am wrong, then I get stopped out. But If get stopped out, and price turns in my original favor, then I put the stop loss in the wrong spot. I just think it comes with practice. I picked a timeframe and sticking with and taking and managing trades and I am watching and questioning everything. Why did price stop there? ohhhhh, because of that trendline from yesterday. ohhhhh because that was high of day yesterday, so sellers take money. Well if price get about this level again, i want some of this money too. While I am waiting, lets see where I wanna be short at. That's it. Everyday, the question is: where do I want to be short and where do I want to be long.
dozu888, good point: The human brain will always and forever out perform the AI. Why? Because its a human brain that programs the AI. I put to you like this. If it was a decision between investing my retirement money in a AI trading system or SP 500 index. I am going with the index every single time.
AI deep learning is not just programming. It's more about having the data and the computing power to grow the neural network. Check out AlphaZero, where the Zero means 'start from scratch'.. it became the strongest chess player in 4 hours...the strongest Go player in days. After KeJie (world #1) lost, he made this comment - 'humans are so unnecessary'. Besides games, AI is already doing better than human in many areas - medical diagnosis, facial recognition, voice recognition, to name a few... agents are already used in the business world as human assistance (for now)... self driving car is another example. And this type of thing usually advances exponentially. DeepMind just has bigger fish to fry now in medical and weather research. this is the 3rd inning... the general public has no idea what is coming... buy every share of FAAMG now and hold on to them.
facebook apple amazon microsoft google. this is a race of data and these guys have the data... and small guys can't compete... that's why we are seeing the QQQ getting more top heavy. China has the BAT - Bidu Alibaba Tencent... they are right behind the US..