What is Quantamental?
The term “quantamental” refers to the investment strategy commonly used by hedge fund managers wherein the traders’ decisions are aided by the results generated using machine learning and artificial intelligence. The combination of quantitative approach and human intervention in the buying and selling of various financial instruments, such as bonds, stocks, derivatives, etc., results in improved performance of the overall portfolio. Effectively, this investment strategy helps in overcoming the flaws of both broken model and human bias.
Quantitative trading primarily refers to computer driven trading approach that involves use of mathematics and finance coupled with trading experience. It helps in building a computer code that can leverage an exploitable market anomaly, such as price arbitrage or momentum trading. These strategies are characterized by holding of large number of securities in order to take advantage of the law of large numbers and its ability to process an infinite amount of data.
On the other hand, the fundamental investment approach implicates the valuation of a business on the basis of its forecasted future cash flows. The forecasting is typically based on expected growth rate and current financial position along with various other factors, which may include both quantifiable and non-quantifiable information, such as economic reports, other analyst valuations, earnings call transcripts, news events, etc.
Now, the word quantamental is the combination of both the approaches mentioned above – quantitative and fundamental. As the name clearly suggests, asset managers intermix various quantitative approaches with traditional fundamentally driven investment strategies to improve their portfolio’s performance in the search of new sources for alpha. Basically, a large pool of potential stocks are assessed using accounting information coupled with statistics and financial econometrics, which necessitates the increased use of machine learning.
There are several steps involved in a quantamental investing process and they can be categorized as below:
- Step 1: Firstly, identify the need or gap in the existing process. Once done, develop an idea to fill the gap to offer a smoother process in the future.
- Step 2: Next, gather as much data as possible in order to build a strong database. Then, analyze the data and draw meaningful insights, such as any particular pattern or trend.
- Step 3: Next, create a model on the basis of the available data and drawn insights. There will be several rounds of iterations to refine the model before the final model comes out.
- Step 4: Next, use the model to quantitatively screen a large universe of potential stocks to identify some investment opportunities.
- Step 5: Next, employ fundamental valuation techniques to the selected assets and come up with the final selection. This where quantitative methods merges with the fundamental approach.
- Step 6: Next, build a portfolio on the basis of finally selected assets and optimize it using the analyst’s experience.
- Step 7: Finally, the system is ready to trade and monitor using algorithms. It becomes a continuous process to analyze the forthcoming data and improve the algorithms as and when required.
Examples of Quantamental
Now, let us take the example of the alpha surprise model that is used by merrill lynch as an instrument for quantamental investing. The wealth management firm developed a new research product- ML Alpha Surprise Model Index. This index helps in tracking the performance of a large number of stocks taken from S&P500 using the Alpha Surprise Model, which is one of the models developed by the US Equity Research Quantitative Strategy group. The ML Alpha Surprise Model Index has been constructed with equal weightage to each stock in the Alpha Surprise Model and these stocks are updated every month.
Over a period of time, the index has outperformed the S&P500 index – higher return at a lower level of risk. During the period from1999 to 2007, the index has generated an annual return of 11.4%vis-à-vis S&P500 that could generate only 1.2%, while the index’s volatility has been lower than that of S&P500.
Risks of Quantamental
Quantitative trading can at times be plagued with the risk of curve fitting, wherein a certain pattern in the available data is used to build the final model. But the model may not be fit for some of the future data points, which will result in the failure of the investment strategy based on the model. Nevertheless, this risk can be mitigated by testing the model prior to finalization using data from outside of the sample.
Some of the major advantages of quantamental are as follows:
- All investments are based on a large volume of the well analyzed data set.
- The amalgamation of quantitative methods and fundamental methods is a step towards a more improved financial market.
- It is a better method of valuation than either the quantitative or fundamental method.
- Historically, investment strategies based on quantamental methods have performed better than most other investment strategies.
Some of the major disadvantages of quantamental are as follows:
- Machine learning algorithms are still an enigma for the human kind and it is very difficult to predict how it will behave under certain unanticipated events. For e.g.: the trillion dollar flash crash on May 6, 2010.
- Machines react purely to triggers as it lacks any kind of intuitive sense, unlike humans.
- Over a period of time, the need of human intervention may gradually go down as machine learning will become more and more advanced. This may result in the risk of higher unemployment.
So, it can be seen that quantamental is an investment strategy wherein statistical methods and mathematical principles go hand in hand with the traditional fundamental investment methods. In short, it offers the benefits of both quantitative and fundamental investing. However, like most technologies, it has its own merits and demerits and it is up to humans to leverage its potential to the maximum.
This is a guide to Quantamental. Here we also discuss the introduction and examples of quantamental along with advantages and disadvantages. You may also have a look at the following articles to learn more –