Man let's be honest the stated phrase (from the Forbes (RUS) article) "B peзyльтaтe зa 15 днeй экcпepимeнтa мoдeльный пopтфeль выpoc нa 2.81% в вaлютe. Этo cooтвeтcтвyeт дoxoднocти 47% гoдoвыx в вaлютe. / As a result in 15 days of an experiment the model portfolio grew by 2.81%. It corresponds to profitability of 47% per annum." is a piece of shit. 2.81% profitability in half a month could be achieved if you play on the growth of bitcoin. what are the guarantees that the robot will show the same profitability during the upcomming period? How can he predict and avoid market failures, thus not loosing my money?
Or even according to the basic logic rules, if everyone or even a big part of investors (we talk about RU zone, right?) start using your software, the stategy will not be unique therefore the promised profitability will be eliminated.
Well when it comes to trading, people love thinking backwards, what if etc. Your phrase could be achieved if you play on the growth of bitcoin is the same if someone says you could earn 7% in one day if you had bought TSLA on Wednesday and would have sold it yesterday or you could earn 24% in two days if you had shorted TEVA 2 days ago etc. Traders and investors know that there is no such thought process in trading. Also, you cant compare crypto markets with traditional markets. Totally different risks, volatility, liquidity.
In addition to that, it was a pilot project with Moscow Exchange. We had never asked about those financial instruments before, most forecasters hadnt been using our app before (totally new users for our ML models). Therefore, the result of this pilot proved the efficiency of the Hybrid Intelligence ecosystem even with a non-trained (on the computer learning models used by us) sample of forecasters.
Regarding your second question, thank you for gentle comment. We know about that issue and know how to solve it. It is described with details in our WP (section 3.1.2 and 4.6). Thanks again!