Be far ahead！DataYes Releases Beneficiary Stocks in Research Reports Data →

Brokerage firms' research reports and gold stocks are the carrier and crystallization of analysts' knowledge and insights on the market, industry and individual stocks, etc., and have important investment research reference value. However, at present, brokerage research reports and gold stocks related analysis still have more space to dig, still have rich Alpha. as the related analysis may be based on the summary of the past, the future outlook, etc., to varying degrees, perhaps already by the market valuation, need to find the potential information gap.

In this regard, DataYes has launched the Beneficiary Stocks in Research Reports Data. This alternative data product is based on research report data, for company and industry research reports, combining investment research logic and AI analysis capabilities such as NLP (Natural Language Processing), focusing on the "Beneficiary" segment text in the research report, through the labeling of the information refined classification, extracting the beneficiary stocks, analyzing the reasons for the benefit and the theme and sentiment, etc. Further mining to find the potential information gap to obtain a more distinctive Alpha. We extract the beneficiary stocks, analyze the reasons, themes and sentiments of the beneficiaries, and then dig further to find the potential information gap to obtain a more distinctive Alpha.

DataYes 's Beneficiary Stocks in Research Reports Data can be applied to event investment and multiple stock picking strategies. Test data show that the research report beneficiary stock data used in investment strategies, the excess annualized return is significant.

Take DataYes 's Beneficiary Stocks in Research Reports Data for rule event backtesting as an example. during the period from August 2020 to September 2023, the non-beneficiary stocks and price rise labeled samples are screened, statistics for the past 30 natural days to meet the conditions of the sample stocks, daily position, from high to low selection of the top 20/50 stocks, with the CSI 500 as the benchmark, with a total cost of buying and selling of 1.6 ‰ (commissions, stamp duty, slippage) ) deduction, position Top20 excess annualized return of 13.8%, position Top50 excess annualized return of 13.1%。

Then take the ?DataYes 's ?Beneficiary Stocks in Research Reports Data used in the long strategy as an example. during the period from October 2018 to November 2023, using the label, whether it has been benefited, sentiment and other characteristics, rolling construction of the top optimization linear model, and predict the next 20 trading days after the release of the research report of the excess return, review the past 60 trading days to determine the selection of the pool of stocks and the value of the factor, the beginning of each month to adjust the position, from the high to the low selection of the top 30 stocks, with a 3‰ total cost of buying and selling, using the CSI 500 as a benchmark, position Top30 annualized return of 18.8%, excess annualized return of 15.9%.