Comparison between computerized and traditional trading

This master’s thesis was born because of the interests of the author in Big Data, Machine Learning and, especially, trading. Trading has evolved substantially in the last decades, mainly because of the influence of these new technologies. The first part of this work, fields of domain, reflects this evolution. The purpose of this part is to give the reader an idea about the fields that are necessary to control if they want to understand the idea behind this research. These fields of domain, Big Data, Artificial Intelligence, Machine Learning and Trading, appear implicitly along the paper. As the reader will see in the chapter about algorithmic trading, machines have algorithms that give them the ability to invest, study the market, find trends…etc. This would not be possible without AI or Machine Learning, as they have to program machines to learn and so complicated processes. Once the reader has understood how these sciences work and what things they are able to do, they will understand the content no matter the level of knowledge they had at the beginning. The next part of the work is about the characteristics that traditional economics has attributed to humans in comparison to the characteristics that behavioral finance establishes now. The reader will understand how finance has seen traditionally human agents, finding that they have been treated as people with a complete rational mentality, independent, with unlimited knowledge and with the purpose of maximizing always their gains. Behavioral economics tries to give a more realistic vision of humans: not always rational, they change their preferences; they are influenced by others…with the aim of adjusting the traditional economic models to these new assumptions about them. A science similar to behavioral economics is behavioral finance, explained in that part too. Behavioral finance explains how humans, with their lacks and abilities, influence financial markets causing a mispricing of securities. This part about humans’ characteristics is mandatory to understand not only the results of the survey conducted with traders, but also the conclusions given at the end. The following section is about the new trends in trading. Nowadays, people tend to invest in passive funds, which are funds that do not try to beat their benchmarks. Two examples of passive investing are index funds and usually ETFs (Exchange Traded Funds). Index funds mimic the behavior of an index, such as the S&P500 or the Ibex35. Copying the behavior of an index is not easy, as the index is not always the same, so a computer is in charge of buying the same stocks as the index and in the same proportion. ETFs are similar, but the difference is that they can be traded like a stock. Another trend, that is also the most important according to the scope of the thesis, is the algorithmic fund. An algorithmic fund uses an algorithm to invest. A team of economists and engineers that use statistics and market behavior data designs this algorithm, building a model that decides in which stocks is more convenient to invest. Presumably, this type of fund should be better, as it uses empirical data. The quantitative analysis conducted in the following part intends to compare this fund to a fund managed by a human. The quantitative analysis consists in selecting a set of computer managed funds and find a similar set of funds managed by humans. This set contains funds that trade different securities in different markets, such as American Small Capitalization Stocks, Chinese Stocks, Commodities, etc. The comparison is made using the annual net return of each of the funds, performance with respect to the benchmark and average and cumulative returns. This part is followed by a qualitative analysis. With the qualitative analysis, the reader can see that what they learned in the part about behavioral economics and behavioral finance is real, as the answers the interviewed traders give show. Again, there is a comparison, but this time between the opinion of quantitative traders and the opinion of traditional traders. The last part is a social responsibility study, that explores the contribution of this work and the subjects treated to the accomplishment of the Sustainable Development Goals, set by the United Nations. Finally, the author gives the conclusion of the study. The conclusion connects all the sections, gives the result of the comparison between the traditional and algorithmic trading and answers to the following questions: Are machines better than humans in the field of financial trading? Will machines substitute traders