Daytrader
During the past few months I have developed an interest on the stock market and the possibilities around it.
As a tech oriented person I couldn't help myself to believe that it's possible to make use of available technologies or ones to be developed to improve the performance on an ecosystem where changes occur in matter of seconds and where sometimes simple human reaction is not fast enough.
Together with a small group of trusted people I joined an initiative to look for a way to just do that. The chosen tool for it was to develop AI algorithms as a way to obtain predictions for possible transactions.
It has been a rough year that has led me to a roller-coaster of impressions about whether it is possible to achieve our goal. Starting with the natural first reaction of skepticism and the following growth of believing it is actually possible to predict some sort of market behavior. Eventually some disillusionment happened and it was time to step back for a second and think about it. Was it human error? Was the system overseeing variables? Or was it involving too much noise? So it was time to revise everything that was done until now.
I believe the problem was caused by a little bit of everything.
The main reflection I have made about it is that the stocks market is not as simple as an offer/demand system. And you probably already knew that. It is way more than that. It is the reflection of human behavior as a society. On it you can perceive joy, excitement, fear, cautiousness, impulse etc.
Events like 9/11 are a sample of how the real world affects the market. During this Black Swan event a lot of people was caught by surprise and was unable to react on it and caused great deal of losses to them. Some others were able to think with their heads cold and obtained significant gains on trading on those moments. Why?
Ironically (I think I'm using the word correctly here) to be a successful daytrader you have to suppress yourself of all emotions when you are trading manually and that's something computers do better than us. But also for computers to be good traders it is necessary that they understand human emotions and how they translate into the markets, and that is certainly something human still "sometimes" do better.
I can thereby conclude that the technological solution should be able to react fast enough by understanding human emotions and behavior but also remaining unaffected by these emotions.
Update:
On a related note to my this post the following article notes how everyday the automation of stock trading is increasingly replacing human intervention on the market decision making. I believe it wont take long until governments start creating regulations to prevent high risk circumstances where the markets could collapse due to systems that might not be well controlled.
As a tech oriented person I couldn't help myself to believe that it's possible to make use of available technologies or ones to be developed to improve the performance on an ecosystem where changes occur in matter of seconds and where sometimes simple human reaction is not fast enough.
Together with a small group of trusted people I joined an initiative to look for a way to just do that. The chosen tool for it was to develop AI algorithms as a way to obtain predictions for possible transactions.
It has been a rough year that has led me to a roller-coaster of impressions about whether it is possible to achieve our goal. Starting with the natural first reaction of skepticism and the following growth of believing it is actually possible to predict some sort of market behavior. Eventually some disillusionment happened and it was time to step back for a second and think about it. Was it human error? Was the system overseeing variables? Or was it involving too much noise? So it was time to revise everything that was done until now.
I believe the problem was caused by a little bit of everything.
The main reflection I have made about it is that the stocks market is not as simple as an offer/demand system. And you probably already knew that. It is way more than that. It is the reflection of human behavior as a society. On it you can perceive joy, excitement, fear, cautiousness, impulse etc.
Events like 9/11 are a sample of how the real world affects the market. During this Black Swan event a lot of people was caught by surprise and was unable to react on it and caused great deal of losses to them. Some others were able to think with their heads cold and obtained significant gains on trading on those moments. Why?
Ironically (I think I'm using the word correctly here) to be a successful daytrader you have to suppress yourself of all emotions when you are trading manually and that's something computers do better than us. But also for computers to be good traders it is necessary that they understand human emotions and how they translate into the markets, and that is certainly something human still "sometimes" do better.
I can thereby conclude that the technological solution should be able to react fast enough by understanding human emotions and behavior but also remaining unaffected by these emotions.
Update:
On a related note to my this post the following article notes how everyday the automation of stock trading is increasingly replacing human intervention on the market decision making. I believe it wont take long until governments start creating regulations to prevent high risk circumstances where the markets could collapse due to systems that might not be well controlled.
Algorithmic stock trading rapidly replacing humans, warns government paper
Algorithmic trading, including high frequency trading (HFT), is rapidly replacing human decision making, according to a government panel which warned that the right regulations need to be introduced to protect stock markets.
Around one third of share trading in the UK is conducted by computers fulfilling commands based on complex algorithms, said the Foresight panel in a working paper published yesterday.
Comments
Post a Comment