Is it possible to entrust investments to robots: a study of the BBC

Several floors of the buildings of the New York and London stock exchanges are currently empty and stored only "for the view". Real trading on them is carried out not by brokers, but by computers, automatically. Approximately three-quarters of transactions on the New York Stock Exchange and Nasdaq are carried out using automated algorithms — computer programs that follow a specific set of rules.

Robo Trading has the strongest influence on the investment world, changing the structure of the global hedge funds and personal managers. BBC News experts examined the advantages and disadvantages of using supercomputers to manage trillions of dollars around the world.
')

“Demystification” of stock trading

The main advantage of introducing new technologies for individual and corporate investors is that they get a powerful tool for managing a balanced portfolio of investments, often at a much lower cost than with the help of brokers or investment funds. If the investor is not able to use the technology on his own, he may delegate business management to a consultant or an intermediary company.

“Robotic” investment companies, such as Betterment and WiseBanyan in the USA, or Nutmeg and MoneyFarm in the UK, are trying to “demystify” the investment process, having innovative tools in the public domain.

WiseBanyan co-founder Vicki Zhou says that its platform allows people to "invest using electronic algorithms and create a diversified portfolio of low-cost securities." According to Zhou, the use of "robo-investors" allows you to reduce the cost of asset management.

Joe Ziemer of Betterment argues that new technologies are helping to reform the process of investing retirement savings. “We look at 40 different parameters - marital status, rental income, pension size, etc. - and, in a matter of seconds, we present online a comprehensive retirement plan. "

A recent report by the UK's Financial Conduct Authority states that online financial advice will “play an important role in driving cost savings.”

This is good news for investors, and bad for investment advisors. Thus, the Royal Bank of Scotland announced a reduction in jobs for 220 managers working with clients in response to the introduction of new technologies.

Satisfying the need for acceleration

Large financial institutions are always looking for advantages over competitors. Information is power, so anyone who owns a large amount of it and can put knowledge into practice faster will win the race for profit. Robo Trading provides companies with this advantage.

Computers can do a lot of trading in seconds, using minor fluctuations in stock prices and stock index changes to maximize profits.

New Jersey's Tradeworx mounts short-wave computer networks and builds repeater towers every 30 miles between Chicago, the futures trading center, and the New York Stock Exchange. This network will transmit financial information 2.3 milliseconds faster than existing fiber optic cables. Even such a small saving of time will be enough for the “robo-trader” to take advantage of the super-high-speed reality of “ flash trading ”.

Greed and Fear

Another significant advantage of the latest supercomputers is emotionlessness. “They don’t panic, they don’t understand things like greed and fear,” said Dr. Michael Halls-Moore, Dr. Michael Halls-Moore, the creator of QuantStart.com, who teaches investment strategy writing.

Robo-managers are getting smarter. With the growth of machine learning opportunities and the enhancement of artificial intelligence, they can analyze more and more news, research and information from social media - plenty of data for potentially learning and self-improvement.

“When information was scarce, people began to collect it and invested on the basis of accumulated knowledge,” says Thomas Wiecki, senior data scientist at Quantopian: “Now we are dealing with such a huge amount of data that a person never will not be able to analyze and automate them. ”Quantopian encourages private investors who are developing their own machine algorithms in the fight for profits in the stock market.

Dr. Eugene Kashdan, a former London algorithmic trader and now a mathematics teacher at the University of Dublin, argues that the technical algorithms used separately will still not be able to reveal a lot of useful information. If, however, the algorithm is used in combination with hundreds of others, the picture “inaccessible to the human eye” is added: the car gives a signal to buy or sell assets.

Getting out of control

Proponents of algorithmic trading argue that it leads to an increase in the liquidity of stock markets and reduces the cost of asset management. Critics of innovative technology believe that it is the cause of "wasting" talents of highly skilled mathematicians and physicists, and destabilizes markets with methods that no one - not even market regulators - are able to understand and describe.

On May 6, 2010, the US stock market experienced an “instant collapse” (or “ flash crash ” - the name by which the incident is known throughout the world, including in Russia ). The value of shares of American companies fell by one trillion dollars. The market “recovered” 36 minutes. And although most of the owners of "long-term" securities did not even notice what had happened, the extraordinary situation alarmed the managers. The US authorities blamed the 36-year-old resident of London, Navinder Sarao, who used available algorithmic software for remote trading on the stock exchange at his parents' house.

The main danger is that “flash crash” can become a frequent occurrence in the world of traders if self-learning robots dominate. The scenario of the collapse of prices "smart" computer for the purchase of cheap stocks and profit after the "recovery" of the market, does not look too implausible.

Stagnation

Some experts believe that the most likely consequence of access to self-learning trading machines to all market data will be the choice of automatics of the same approach to work and, consequently, to market stagnation. Trading volumes will decrease with the spread - the difference between the purchase and sale price of securities. “The best and worst scenarios are pretty close to each other,” says Dr. Eugene Kashdan.

However, there are those who believe the markets will never reach this point, because the world is too complicated, and no algorithm will ever be able to predict the future. “Each of us acknowledges that this is not possible,” says Quantopian hedge fund CEO John Fawcett: “But the problem is too important not to think about it.”