Mechanism stock selection logic gene mutation quantification investment Internet digging bokhary data

21st Century Capital Management Ningxia Shenzhen Report

Using the Internet financial data to build stock selection model is becoming a new trend in a-share market.

October 20, the GF Fund and Baidu company developed the Guangdong China card Baidu Hundred Strategy 100 Index fund will launch, this is the industry's first tracking the internet genetic index of the index-type fund products.

There are also Southern fund Management Limited, which plans to launch the Internet Gene Index products.

"The Fund Product program, which tracks the Big Data Index, will be formally declared to the regulator in the near future and is expected to issue a fund product following the I100 index this year." "South Fund product Development Department deputy director Ruli revealed."

According to the 21st Century Economic report reporter to understand, in addition to the GF, the South fund, the industry also has the fund's quantitative investment department and the Special Division in the study of "large data" quantitative strategy.

In addition, the brokerage Institute's financial engineering team is also digging "big data" investment opportunities. The industry representatives for the Yangtze River Securities financial engineering team, they early in the beginning of the year, according to Sina Financial news data to build a news stock model.

The glamour of Big data

It is becoming a new trend of the fund industry to use the big data of Internet to excavate the excess income opportunity of the stock market. And the quicker action is the GF Fund and the South Fund of the two fund companies.

Among them, the GF fund United Baidu Company, the China Certificate Index company Development Hundred Hair 100 Index, the South fund is working together Sina Finance, the deep proof information company has launched the i100 index and the i300 index.

The incentive for fund companies to scramble to develop large data indices is to outperform existing index funds based on a combination of large data filters.

According to the data from the China Certificate Index company, the cumulative yield of the 100 index reached 545% from 2009 to 2014 June 30. In the same period, the yield of the 500, China certificate and the Shanghai-Shenzhen 300 index were 102%, 56% and 19% respectively.

There may be doubts about the profitability of historical gains derived from the simulation of model samples. and the 100 index products in the model sample of the firm data, the same big win the mainstream index.

Jifeng, a researcher of the quantitative Investment Department of GF Fund, told reporters that June 20, the Hundred 100 index into the operating stage of the real disk. From this to October 8, the cumulative yield of the 100 index reached 43.33%. In the same period, the Shanghai and Shenzhen 300 Index, the Shanghai index and the gem are only 16.52%, 17.74% and 17.94% respectively.

The historical yield of weight index such as i100 and i300, also highlights the advantage of large data.

From January 29, 2010 to July 31, 2014, the cumulative yield of i100 and i300 respectively reached 222.4% and 141.58%, which is much higher than the cumulative return of the gem index 34.45%, and the cumulative yield of the SME Board Index-8.95%.

South Fund Quantified investment director Liu Jiping introduced, the southern Sina Large data index since August officially entered the model outside the operation. Among them, i100 and other weights index in August and September, the monthly yield rate was 10.93%, 15.63% respectively. This means that the index has a yield of 26.56% per cent over a two-month period.

The research team of brokerages is also tapping into the investment opportunities brought about by "big data". Among them, the Yangtze Securities financial engineering team since the beginning of the establishment of the news stock selection model, since January 12 to start tracking outside the sample.

Fan Xinting, director of the Yangtze Securities and financial Engineering, released a study report, January 22 to August 15, the news stock selection model cumulative absolute yield reached 52.45%, beyond the Shanghai and Shenzhen 300 index 39.44%, beyond the 500 index 30%.

China Merchants Securities Financial Engineering Senior analyst Xia Yangyang use Shenzhen Interactive easily disclosed research information, the establishment of the SME Board research portfolio. The results of the firm tracking show that from the beginning to September 30, the cumulative yield of the research portfolio was 29.31%, and the SME index was 14.91%, surpassing the gem index of 8.53%.

Introduction of the Internet gene

Whether it is the stock selection model of the Changjiang Securities and China Merchants Securities, or the large data index launched by the GF Fund and the South Fund, the common point is to introduce a large number of users ' search data as the stock-picking factor.

"The principle of the Hundred 100 Index is based on the traditional index model, to increase the search factor of large data Baidu, the equivalent of the upcoming behavior of investors into the index indicators." Lu Zhiming, general manager of the Quantity Investment Department of GF Fund, said.

Hundreds of 100 of the index of the participating developers, Jifeng, a researcher of the Quantity Investment Department of GF Fund, introduced that the model of 100 is also called the investor's comprehensive emotion model, which is equivalent to a large data melting pot, which makes the analysis of several kinds of data such as financial factor, momentum factor, search factor and so on.

"Using the data of Baidu platform, we calculate the total search amount and search increment of the last one months for the stock of the sample space respectively, which is counted as total factor and increment factor respectively." "Jifeng Introduction, Baidu Financial data, in a sense is a measure of investor sentiment, one of the quantitative methods can be used to reflect the investment will or expectations of investors."

"The innovation of the Hundred 100 Index is that the Internet search volume has become the influential factor of the choice of constituent stocks, and the construction of the index utilizes the strong resources of Baidu in the search field and has significant internet genes." Said Liao, a researcher at the Research Center for the Fund.

The information disclosed by the China Certificate Index Company's website shows that the sampling method of the 100 index is in descending order according to the comprehensive score of comprehensive financial factor, comprehensive momentum factor and search factor, and selects the top 100 stocks as the constituent stocks of the Hundred 100 Index.

Among them, the comprehensive financial factors include 8 financial factors, such as net assets yield, asset yield and earnings growth rate per share. The synthetic momentum factor is to calculate the price yield and volatility of stock in the last one months, and get the momentum index after the risk adjustment. In accordance with their size from high to low, ranked in the top 30% is the momentum factor, and the latter 30% is recorded as reversal factor.

South Sina Financial Large data sampling method is also introduced into the domestic first door household Sina network of large data.

The stock selection factor of "I Index" includes financial factor, large data factor and market driving factor. Sboll, director of the Southern Fund Equity Investment Division, said the big data factors covered the page clicks of Sina Finance, the positive and negative articles of Weibo, and the impact of news reports.

In addition, financial factors include the latest P/E ratio PE, ROE, annual operating income and net profit growth rate, and their change forecasts. The market driving factor refers to the stock turnover rate, fluctuation rate, price change rate, volume change ratio in the near future.

Long before the South Fund, the financial engineering team of Changjiang Securities used Sina Financial data to build a news stock-picking model.

"We have selected the top 50 of the list of news-focused, fastest-growing weekly hits by the data on Sina finance, culling companies with bad news and building a stock-picking portfolio of 50 stocks. "Fan Xinting points out that the core of the news stock selection model is to select a stock-building portfolio with a greater increase in recent clicks in order to find the expected changes brought about by the news heat."

Although the introduction of the Internet financial large data as a stock-picking factor, but the composition of different institutions have a significant difference.

According to the information disclosed by the China Certificate Index company, the sample stocks in each market plate are relatively dispersed in the first 100 constituent stocks of the 100 Index. Among them, the number of SME stock, up to 39. The second is the Shanghai board 31, the Shenzhen board and the gem boards each accounted for 25 and 5.

From the industry distribution, the 100 index of the constituent stocks of the top five industries are industry, optional consumption, raw materials, financial real estate, information technology, the total proportion of the weight of the index reached 85%. Compared with the Shanghai and Shenzhen 300 index, its financial real estate allocation ratio is less, information technology and optional consumption and so on obviously match.

According to financial factors, large data factors and market driving factors, the constituent stocks selected by I index are mainly distributed in four industries, such as raw materials, information technology, industry and optional consumption. Among them, the i100 Index, i300 index in the above four sectors of the weight accounted for 73% and 71% respectively.

Subversion of stock-picking logic

When fund companies and brokerages study the large data of internet finance as stock-picking factor, it represents the restructuring of stock-picking logic in index investment by asset management agencies.

Traditional indices rely on traditional factors such as market size, turnover, finance and valuation. Its biggest drawback is that it uses data from the past 3 months or 6 months to predict future earnings, and parameters are heavily dependent on historical data. "Lu Zhiming pointed out that the use of Baidu data into the investors in the investment decision before the behavior of the law, the future of the market investment law has a certain role in predicting the effect of better than the traditional data from the historical factor data.

In contrast to the traditional index, the idea of the 100 index is to jump out of the industry, the plate limit, looking for the opportunity of excess income in the whole market.

"We want to try to find a hot spot in the market for the next one months, through financial data, momentum data and big data from Baidu," he said. "Jifeng that the 100 index stock model is characterized by its choice of stocks is the future market or industry rotation hot spot, and the fundamental good, the future has a certain growth space of value-oriented stocks."

The Big Data Index, launched by the South Fund and Sina, also uses the same idea.

"The use of large data investment to market breadth to excess income." "South Fund equity investment director Sboll, according to wind data statistics found that from 2005 to 2014, even if the overall market average yield is negative, still can find out 30% of the market shares.

Take the beginning of 2014 years to August 22 as an example, wind full a yield is 14.12%, beyond the market 30% of the number of shares to 535, accounting for the proportion of the total market share is 22%. Among them, the average yield of this group of shares reached 74.63%.

In the industry's view, the fund's Quantitative investment department has introduced large data on the Internet, with the aim of digging out the target of such an average return on the market. Based on this, compared with the traditional index semi-annual sample stocks, the Hundred 100 Index and the i100, i300 index of the sample stocks of the unit is set to the frequency of each month.

"About 50% of the constituent stocks will adjust every month." Jifeng, the 100 index is being sampled in Friday for the third week of the month. Three samples have been adjusted since June 20. Of these, about half of the constituent stocks are replaced each month.

Jifeng Frankly, compared to other indices, the 100 index component stocks of the adjustment ratio is relatively high. However, he believes that the index sample adjusted frequency, far more than the traditional index, the purpose is to better grasp the market rotation.

Shenzhen Securities Information Co., Ltd. according to the data provided by Dr Shiping, from July 1, 2013 to June 1, 2014, the i100 index averaged 41 shares per adjustment in nearly 12 samples of the southern Sina large data, and the I300 index averaged 95 sample stocks per adjustment, The monthly average adjustment ratio was 43.5% and 34.5% respectively.

In the industry's view, the past two years, the domestic market has a clear theme of the characteristics of the wheel movement. The strategy index based on "big data" is very suitable for the market of the topic wheel movement.

However, some research institutions have suggested that the introduction of large data models can withstand the test, still need time to verify.

"The establishment of the 100 index and I coefficient index, while considering the value of stock, also actively grasp the focus of the market and public opinion, such a starting point for good performance is beneficial." "Liao that the quantitative model used to build the internet gene is reasonable enough to give the answer.