ITI_Capital August 16, 2016 at 12:34

Think financial models: Who are quanta and how to become them

Transfer

Turing Finance blog editor and, concurrently, hedge fund financial analyst Stuart Reed has published a practical guide for those who see their future in the financial market based on their experience in the profession. He promises to tell not about how to become a quantum, but how to be it in any of the sectors of the financial industry in which you currently work. In his opinion, this is not even a question of learning, but rather a question of ideology. We present to your attention an adapted translation of this material.

Quantitative methods or quantitative analysis in the finance industry is a science, but not a profession, says Reed. This means that knowing how to become a quantum adds nothing to your skills. You need to know how to be him. In the second case, we are talking about the principles and ideology underlying the quantitative analysis. In the first version, it’s about showing some practical activity in order to get a job where the word “quantum” is in the description.

At the basic level, being a quantum does not mean the ability to derive sophisticated formulas for assessing the possibilities of a strange stochastic model. This does not mean to create, train and test trading strategies using the statistical regression model. To be a quantum means to believe that scientific models are suitable for a general analysis of financial markets.

Such an ideology gained weight in evaluating derivatives, then switched to risk management, asset management and stock trading. Probably in the next decade we will be able to observe the spread of this approach in the field of corporate finance, in venture and banking investments.

Philosophy of science

In his final year in computer science, the author, along with the rest of the students, spent a year creating a module called “research methodologies”. It was based on two textbooks: “Philosophy of science: from problems to theory” and “Philosophy of science: from explanation to justification”. Both are written by Mario Bunge. Despite all the teacher’s attempts to make this subject as boring as possible, the author liked to study philosophy. Many ideas of the module turned out to be in demand in further practice.

Actually, the subject “philosophy of science” itself tells about the logical foundations, methods and conclusions of science. Simply put, he is trying to answer the questions: what is science? How it works? What are the limits of scientific knowledge? History has given many answers to these questions. Often the opposite. They can be reduced toseveral directions : realism, empiricism, instrumentalism and induction, logical positivism and falsifiability.

Realism versus Empiricism

Realism is convinced that science deals with the real world in its research, and not with its construction, which is given only in sensations. Empiricism, on the contrary, proceeds from the proposition that we know everything that we know through the senses. Therefore, our knowledge is temporary and limited.

The main argument in favor of realism may be the accuracy of scientific theories. If we have a theory accurate enough to be able to predict the future state of the system with it, then we describe the real world. This is more true for the physical sciences. In the financial world, any theory will be limited in its ability to predict and subject to distortion.

Deduction, induction and abduction

In the philosophy of science, we distinguish several forms of logical reasoning.

Deduction means inferring an inference, based on one or more assumptions, to one consistent conclusion. It is often used in mathematics and formal methods of computer science.
Abduction is a somewhat simplified form of deduction, which claims that obtained through the latter conclusions may not be true as necessary. But, in any case, this is the best we can do based on the available data and assumptions.
Induction draws conclusions on the basis of one or more strong premises, based on evidence and observation, to obtain probabilistic truth that can be verified by falsification.

The disadvantage of the latter method is that any conclusions here will be unreliable. This can be demonstrated by the well-known example of a black swan. As long as all the swans that come across our path are white, we can be guided by the statement "all swans are white." The first black swan, the appearance of which the model cannot predict, will refute this truth. Nassim Taleb considers this example in more detail in his writings .

Quantitative market theories are built on induction through empirical observation. Such theories are easily refuted in the presence of conflicting data. They may be slim on paper, but in practice they are most often the subject of faith of their adherents who miss facts that do not fit into the model.

The reader is probably already wondering how the philosophy of science helps to become a quantum? Her ideology is important for understanding the limitations of our knowledge of financial markets. No matter how deep financial theories are rooted in empirical evidence, seem infallible, they are fragile in nature. These include all the popular premises of quantitative models: normal distribution returns, linearity, stationarity, random walk hypotheses and market efficiency.

Scientific method

The scientific method helps to make thinking clearer and more rigorous and increase the testability of the model and the verifiability of its underlying hypotheses.

The scientific method is a continuous process, including systematic observations, quantitative measurements, experiment, obtaining hypotheses, testing hypotheses and improving them. The author suggests going all the way to the scientific method using the example of random walk theory.

Observations

In the context of financial markets, we usually extract useful information from books, articles, media, and even sometimes, reading good (or not so) blogs on a given topic. What do we find in them? In many academic publications, there is a statement that the market behaves randomly, stochastic processes operate in it.

It is important here to start asking questions. The first questions that will help us somehow evaluate this assumption may be the simplest: who, what, when, and why? Let us try to look at the random walk hypothesis from this point of view .

To whom, in terms of the market, is this theory applicable? Do all markets behave identically randomly? How does randomness relate to liquidity and other parameters of relative efficiency?
What, what forces make the market random? Does efficiency really lead to randomness? If the market is random, then it makes no sense for its agents to compete on it. That is, if efficiency is removed, will the market remain random?
Where is the market random? Are there places in the world where it is slightly less random, for example, in developing countries? If, yes, then what about the differences in the dynamics of this process?
When do markets become random? That is, are they random throughout the entire time or interrupted by modes where there is no randomness? If so, does this mean that at such moments it becomes possible to benefit from such a state?
Why are markets random? What processes and forces keep them in a state of randomness?
How can you measure market performance and test the random walk hypothesis? Can we simulate an efficient economy using the base agent model and check if the price disclosure mechanism is random?

Someone once made an answer to some of these questions. Therefore, it is critical to keep abreast of previous studies. After formulating your own questions and reading literature on the topic, competent and correct ideas will start to arise in your head, which should eventually form a testable scientific hypothesis.

Hypothesis formation

A hypothesis is a declarative statement that justifies the relationship between a set of variables. A good hypothesis should be concise, verifiable, taking into account all previous accumulated research experience. Take, for example, the following interesting idea that one of the regular readers of his blog sent to the author.

Market returns are a coincidence, as the market is able to adapt quickly to get rid of any weaknesses.

Thought is good. But the hypothesis itself is so-so. Many terms are not defined, too much is mixed in one pile, and it is not so easy to check. Let's try to break this hypothesis into several separate ones.

Hypothesis 1 . Market movements (up or down) are indistinguishable from Martin-Löf binary random sequences.

The author’s analysis of the random walk hypothesis led him to conclude that the market is not random, at least in such harsh terms. Following this, many other questions arise that require additional research. Therefore, this new hypothesis can be divided into two parts.

Hypothesis 2 . Market efficiency, temporary availability to all participants of any information, forces the market to develop by chance.

In one of the following materials, the author promises to tell how to test this hypothesis for truth. In the meantime, those who are interested can read two articles on this subject: “Risk Aversion and the Martingale Property of Market Prices”, Le Roy, 1973, and “Asset Prices in the Exchange Economy,” Lucas, 1978. (for reference, both materials are for some reason unavailable - approx. translator). In them, researchers are trying to figure out how access to information of rational market agents affects the random distribution of prices. In general terms, the conclusion is this: free access to data may lead to random market movements, but may not. That is, efficiency does not mean randomness.

Further, all this leads us to the third hypothesis, which is studied relatively rarely, due to the fact that it is difficult to verify.

Hypothesis 3 . Regardless of whether the market develops by chance or not, all the opportunities to make money on it are washed away too quickly for investors to build an ongoing effective strategy.

This hypothesis is not too concise and declarative. It is quite difficult to refute or confirm. That is, in all respects, it is not as elegant as the previous ones. Therefore, at this stage, it can be neglected.

Development of test forecasts

In order to build a forecast, you first need to determine whether your hypothesis is true. Then you should understand what values to include in the forecast. For example, if we take the first hypothesis as the base one, then market returns (up or down) can be checked and calculated using a set of statistical tests NIST. Their behavior should correspond to the Martin-Löf binary sequence. In one of the previous posts, the author has already done this operation using a pseudo-random number generator called the Mersenne Whirlwind. It turned out that this hypothesis is incorrect.

Many people make the same mistake: they are sure that research is all about one objectivity. In fact, it is not so important whether the hypothesis is true or false. It is important that in any situation, we bring new information to scientific knowledge on a specific problem.

In order to test the second hypothesis, one will have to go one step further: create a model of the base agent in which efficiency will be guaranteed. From this model, we then extract the return sequences and test them for randomness. About the models will be described below.

Prediction data collection

The name of this stage speaks for itself. The only thing worth warning: data for verification must be taken from the real world (empirical data). In extreme cases, they can come from a model corresponding to a hypothesis. Both approaches have their pros and cons. With empirical data, a lot depends on how you measure, collect, and store them. Simulated data suggests that the execution of the model was done correctly.

Clarification or refutation of a hypothesis

Based on the data collected, we can answer the question of whether the predictions were correct and whether the evidence supported our hypothesis. It is important to note that with a positive option, it is only about supporting the hypothesis. We cannot prove that it is true. In the second case, the hypothesis is considered false.

Creating a General Theory

So, we have collected a sufficient number of hypotheses, carefully tested them. Now we can collect from them one general theory. For example, portfolio theory took decades of research into the relationship between risk and profit before it was accepted by the scientific community. At the time of the publication of the doctoral dissertation by Markowitz, practically no one believed in portfolio theory. They even wanted to refuse to award him the degree of Doctor of Economics.

The story, in truth, is very instructive. If your idea is unpopular, it does not mean that it is incorrect. The community of financiers is very conservative. Outdated ideas and approaches are defended here with almost religious pathos and seriousness. Quantum’s skills include the ability to look at this world objectively and constantly to struggle with myths about financial markets. Forget consensus, search for truth.

Think in models

How to become a quantum? Design your ideas in a model. Then use these models to streamline your thinking, test and substantiate your ideas, and reveal hidden patterns.

A model is a representation of individual objects or processes that exist in the real world. To build models, quanta use the methods of mathematics and computer science. A quantum, for example, can assemble a model of risks associated with a specific portfolio of assets. Why are models highlighted? Moreover, some believe that this type of thinking was at the heart of the 2008 financial crisis.

Models help us think more clearly.

Developing an idea to the level of a suitable model made in code or written in mathematical formulas does not matter, it makes it more clear to see the meaning, advantages and disadvantages of the idea itself. We look at speculative things in terms of inputs, outputs, and technological processes. Through the model and processes, you can detect missing parts and correct inaccuracies.

Models are testable, no intuition

As soon as an idea has been systematized and encoded into a model, it becomes verifiable. In the end, we can see how the idea corresponds to the real alignment of things. Take, for example, the stochastic model of Brownian motion as applied to the securities market. How does it compare with the real world? Do they take markets into account? Does it take into account periods of high and low volatility? The answer, of course, is no. All this forces us to develop a better stochastic model: the Merton model of diffusion jumps and the Heston model of stochastic volatility.

Another common option for checking the model is to see how it behaves on historical data. Take a conventional investment strategy of biased value. As long as it is based on the intuition of the broker, it is impossible to verify whether it could bring good returns in the past. You can only believe the words of the financial manager that he is good in his field.

Models help you find hidden patterns.

Forget for a while about searching for patterns using machine learning and neural networks. Even obsolete simple models can find hidden patterns and open up a new understanding of familiar things.

Take an example from another area. How many people do you think a particular city should be racist for racial segregation to reach 80%? Scientists have found that it is enough that 30% of people are racists for the emergence of racial segregation in a relatively isolated society. All this can be calculated using the old Schelling segregation model. And such examples, when the model reveals to us new knowledge about familiar things, are mass. For those who are interested, the author advises to take a course in understanding models at Coursera .

Mathematical or computer models help us move away from speculative constructions and remove the cognitive load on an individual person. Simply put, they make us smarter. Much smarter.

Conclusion

Quantitative finance is an ideology, and being a quantum means much more than being just a mathematician or knowing how to write code. This is a story about the commitment to the scientific method and the ability to apply it to study financial markets in general. Given this, the author gives the only advice to those who intend to become a quantum: just be it, regardless of the name of your position. There is no reason that fundamental principles and quantitative methods could not be applied to other areas of financial services or even to non-financial companies. It is likely that in ten years both quantitative banking investment and quantitative venture capitalism will become familiar. Even if you can just correctly paint the principles of this ideology at your interview, this will already be a big plus.

Other financial and stock market related materials from ITI Capital :

Tags: