General quantitative trading platforms, such as Jukuan and Mikuang, have a research module in the form of a notebook. So what is the role of this quantitative research?
Quantitative research data sources:
The quantitative research module data mainly includes three major parts of data, one part is real-time data, that is, the two major stock exchanges tick-level real-time market data; the other is historical market data, the current quantitative platform contains daily and minute-line market data from 2005 to the present ; Another part is fundamental data, such as the company's financial data from 2005 to the present. In addition, there are other data such as indexes, industries, and sectors.
The quantitative research data sources include two aspects. The real-time data is obtained from the Hang Seng Jingteng H5 market server, the historical data (including historical market and historical fundamental data, etc.) is the data of our local quantitative server, and the data of our local quantitative server is Go to the Hang Seng server to download incremental data every day after the market.
The purpose of quantitative research:
A quantitative strategy needs to go through several steps: 1. First start with an inspiration, then obtain market data or fundamental data, and clean and analyze these data; 2. Realize your own ideas through the establishment of mathematical modeling; 3. Then it is transformed into strategy code; 4. Verify the strategy is good or bad on the backtest platform; 5. If the effect is not good, adjust the strategy; 6. After adjustment, you can run the strategy in the trading module.
Quantitative research is mainly the first and second steps, that is, the part of cleaning, analyzing, and establishing mathematical models of stock market and fundamental data.
Introduction to the functions supported by the quantitative research module:
- We provide a standardized jupyter notebook cloud research platform, which provides a python cloud operating environment, on which you can run python code, users do not need to install the python operating environment on their local computer. In short, the research module is a web page embedded in the quantization platform . Users can directly write python code and run the code on this web page, and the running results of the code will also be displayed below the code block. And the code can be run line by line, so that every time we run a line of code, we can see the execution result of the code,
- Some functions are built-in functions of Strategy Platform, such as the function get_price to obtain market data. Research platforms can also call these built-in functions to obtain data for analysis and establish mathematical modeling.
- In addition to using built-in functions, you can also use some third-party libraries, such as the most popular financial data analysis library pandas, machine learning library sklearn, etc.
The similarities and differences between the quantitative research module and the backtest module:
Same point:
1. Can run python code
difference:
- The code of the backtest module runs in the whole section, and some functions are essential, such as the initialize function, which is used to initialize data, such as setting slippage and subscribing to stock market information;
- The code of the research module is run line by line, which is convenient for observing the results of the code.
- Some functions can only be used by the backtest module, such as the get_history function that can only be used by the backtest module to obtain historical market information. The parameters of this function do not need to fill in the securities code, because the backtest module has already subscribed to securities. The research module does not have the initialize function to subscribe to the security code, so it can only use the get_price function to get the historical quotation. The parameters of this function need to be passed in the security code.