The BridgeWise team with the short dog. Image: BridgeWise
The system built by the Israeli startup BridgeWise (formerly Deshe Analytics) performs a daily in-depth analysis of the financial report data of all stocks in the world for banks and investors, and rates for them the probability of each company providing an excess return over similar stocks, i.e. its reference group. In this way, the startup can give the banks a relatively simple analysis to read each stock at the click of a button, and understand what is happening in all the markets in the world and act accordingly.
In a conversation with Gigtime The company’s CEO and one of the founders, Gabriel Diament, explains that the company’s system is based on in-depth financial research, in the framework of which they took significant knowledge bases from the world of economics. They combined the theory of the efficient market, stochastic pricing of shares, value investments and post-earning drift to produce an algorithm that can To predict well the performance of the stock for the following period of time. “From among 252 parameters we created, based on researches, we select the most significant parameters and weight them by supervised algorithms, but this is only the last step in the process. The weighing process happens twice a day and relies on all the data until the weighing is re-weighed.”
According to him, behind the scenes there are 66 algorithms that divide the world of public companies into segments, when the information is collected on a daily basis and from which 252 seasonally neutral factors are calculated based on the raw number. “We throw these numbers on a unique distribution in front of the other companies in the segment and in this process we run a lasso for the selection of the parameters and then a supervised algorithm for weighting and performance prediction,” added Diamant.
What makes your product more successful compared to tools that are already available to the public and provide investment recommendations?
“Our technology combines two technologies: the first is an analysis technology that knows how to analyze each reference group in the market in a specific, high-quality and in-depth way. The second is our ability to translate the analysis into understandable and simple human language, in any language (professional financial language, but understandable) and is something that will be difficult to compete with in it (but always possible of course). The combination of the two creates a significant Barrier to Copy, when in practice it will pay off for an organization that wants to use the product to pay us for the service than to develop it internally for tens of millions of dollars. In practice, the specified language models make information accessible, not knowledge. They know how to collect information Raw and not analyze the information and provide conclusions.”
Recruiting from Dubey Frances
Today (Tuesday) the startup reports the completion of a seed round of 13 million dollars led by the Group11 fund of the investor Dobby Frances, with the participation of the investment arm of the Brazilian Stock Exchange (L4VB), the investment fund Mangrove Capital, Jeffrey Keel, former president of National Republic Bank; Howard Morgan, founder of the Renaissance hedge fund; and David Majerman, Renaissance’s Director of Quantitative Investments. The company’s clients are banks, investment houses and trading platforms, including Bank Hapoalim, the International Bank, Psagot, Excellence, Meitav Dash, IBI and Interactive.