Most fintech companies offer better interfaces for handling money (bank accounts, loans, payments, etc.), but quite a few startups are also offering a solution for a much older need than how to make money.
Israel company I Know First, managed by CEO Yaron Golgher, is one of these. The company's main product is an algorithm that provides a forecast for three thousand different investment instruments, including shares, commodities, interest rates, foreign currency, exchange traded funds (ETFs), and global indices.
"The algorithm rates all the investment instruments, and singles out investment opportunities in the capital market on a daily basis, according to the pricing anomaly it finds," Golgher says in a "Globes" interview.
"Globes": What information do you use?
Golgher: "The algorithm is self-learning. It is based on purely quantitative values, not reading news or any kind of analysis. There is no human factor here. The algorithm uses artificial intelligence, an area in which huge companies like Apple Computers, Google, and Facebook have recently been making massive investments. The algorithm was developed by our development team, headed by cofounder and CTO Dr. Lipa Roitman. Roitman is a scientist from the Weizmann Institute of Science with over 20 years of experience in the specific field of artificial intelligence, machine learning, and algorithms. The forecast is given for a period of time. What is interesting is that for every forecast, the algorithm also assigns a probability that the forecast will be fulfilled. Every customer can receive a forecast according to his investment preferences. For example, a person investing in technology shares can receive the best opportunities in this segment, a customer investing in commodities will receive the best opportunities in the commodities market, etc."
"12 new countries"
Does it work?
"We have expanded this year to 12 new countries, including the US and Europe, with an emphasis on Italy and France, and Russia, too. We work with Latin America, especially Brazil. The business model is based on access to the algorithm to a varying extent, according to the customer's size."
Are your customers banks? Private individuals?
"We have institutional customers who receive broader access to information, including personalization of the algorithm according to the portfolio they manage, and also private customers looking for more advanced tools for spotting opportunities in the market. They're looking for tools that are not necessarily technical analysis or based on news or human elements. This is an independent tool that rates the various investment instruments. In addition, in the coming months we'll launch a hedge fund with a financial agency in Israel, and towards June, we plan to launch a hedge fund in the US and with an investment house in Brazil. Another point is that the algorithm can also be used in environments other than the capital market, given the right information. The only condition is that the system has to receive many historical points of information for the purpose of the learning and forecasting process. In today's world, there is a lot of information that has accumulated in various companies; the main challenge is to be able to spot future trends using the information - to identify the significant information, and to filter out irrelevant information. Our algorithm can do all of these things.
What else can you predict?
"Right now, there's a financial agency interested in producing credit forecasts for customers according to their history and characteristics. Insurance companies are also interested in identifying client risks in the policy production stage, and they of course have historical information that can be analyzed. The algorithm has to learn. It requires 'clean' historical information in order to supply a forecast with a level of significance. The algorithm can also be used as a tool for personalized medicine in order to predict health problems by monitoring the accumulated information (on a smart watch, for example), or to predict future road traffic and daily demand for energy and electricity.
"Dr. Roitman started out in chemistry. He worked at Dow Chemicals in the US, where he had to develop a model for forecasting chemical processes with many variables that cannot be predicted in simple ways. From there, we started looking for many information environments, and came to the capital market, in which there is information every second."
"To identify the trend"
I Know First has offices in the Tel Aviv Port, where the company has 11 employees. "We also have overseas employees, who help us develop these channels," Golgher says. "There is also interest in the algorithm in Far Eastern countries: China and Japan. What's nice is that if a customer is interested in getting a forecast for the Japanese stock exchange or the local stock exchange in China, he can certainly use the algorithm after it has been individually adapted. Obviously, we're now focusing on the US and Europe, but growth will be in the direction of individual adaptation for each geographic region.
"We have grown substantially. Since the initial financing round, we have always made a profit. We didn't need to raise money at all. The initial investment was $1.2 million, and we have raised no money since then."
The $64,000 question is how accurate your forecasts are.
"We give forecasts for different time periods. The longer the time period, the greater the statistical significance and the greater the accuracy. The algorithm is able to identify the trend, and to filter out the background noise. The longest forecast is for a year, but we also have a forecast for three month, and even for three days. The accuracy varies for different investment instruments. Some investment instruments are easier to predict, some less so. The customer knows this indicator in advance. For example, we tell him that our forecasting ability is better for the Apple Computers share than for the Facebook share.
"Indices usually have higher levels of significance, because the risk is spread. In principle, the algorithm performs initial filtering for all three thousand investment instruments, and selects the ones that can be predicted. It filters out the assets with lower levels of significance."
"A fluctuating market"
Let's talk about the asset you are best able to forecast for the longest time period. How accurate is your forecast?
"We checked this for two years as part of our preparations for launching the hedge fund. We built a portfolio cosisting of 60% shares, 30% interest rates, and 10% foreign currency. For a two-year period, this portfolio generated a 92% return, while the S&P index generated something like 6%. Our accuracy percentage varies. It's around 70% for foreign currency."
Could this be a self-fulfilling prophecy? Could you be making a certain asset go up in price by recommending that customers buy it?
"The algorithm focuses on assets worth more than $2 billion. We filter out the small assets. Theoretically, you're right. If millions of people are using the algorithm, then it’s true, but the daily turnover in the shares we analyze is very, very high, so it's hard to move these shares."
If you've really solved the problem, why do you have 11 employees, and not 11,000?
"That's an excellent question. High-quality technology doesn't necessarily require thousands of employees. We're not saying, 'We have a machine that will make you rich if you buy it.' It isn't magic; it's a very advanced decision- supporting statistical model. We don't manage money for customers; we're just advisors. This is an advanced tool for making decisions in the capital market. We only provide the technology to the user."
A hedge fund using the I Know First technology will be launched soon. Golgher believes that the fund will manage $150-200 million. "In the first stage, there will be less, but the goal, based on the performance in the first year, is to reach the sum I mentioned. We cooperated with a large entity I don't want to mention by name. We worked with them for two years to build the optimal strategy, not just for the return, but also for the standard deviation (reducing the risk, I.R.), and we arrived at 60% shares, 30% interest rates, and 10% foreign currency."
"Spotting opportunities even in a bear market"
Does it matter to you if the market is going up or down?
"Not necessarily. The algorithm spots investment opportunities in both a bull and a bear market. The algorithm also gives an indication whether the market is a bull or a bear. Furthermore, because we're looking at all money transactions in the market, as soon as the money is taken out of shares, the algorithm assumes that it's entering other elements in the market. The algorithm is able to identify the formations if shares are a bear market, and tell what's going to go up - which shares are correlated with a given asset. For example, if the price of oil is falling, aviation companies will go up and oil companies will go down.
I can predict that even without an algorithm.
(Laughs) "The thing is that the algorithm never read about these relationships in the press, and doesn't know that the ticker symbol XOM stands for ExxonMobil. It studies relationships between various assets. We're careful not to give it any human information. For the S&P 500, it predicts both the index itself and all its elements, so we have two independent forecasts here that are more statistically significant. In essence, this is a support tool that provides more sophisticated investors with a relative advantage."
There are thousands of such businesses around the world.
"No one knows what the others have. It's a professional secret. What's nice about us is that we also given private investors access to the information. In this sphere, it's usually banks that manage models like this. We don't deny this access to the general public."
How much does it cost?
"A monthly subscription with no obligation costs $200 a month. It varies according to the extent of the access to the algorithm reports."
Published by Globes [online], Israel business news - www.globes-online.com - on March 8, 2016
© Copyright of Globes Publisher Itonut (1983) Ltd. 2016