Qlib is an AI-oriented quantitative investment platform that aims to realize the potential, empower research, and create value using AI technologies in quantitative investment, from exploring ideas to implementing productions. Qlib supports diverse machine learning modeling paradigms. including supervised learning, market dynamics modeling, and RL.
Yes, I read the docs about adjusted price data and how to dump my own csv to bins. But the problem is, the price I got in qlib is far diferent from actual price:
for stock SH600519, use code below:
qdl = QlibDataLoader(config=(['$close', '$high'],['close', 'high']))
qdl.load(instruments=['SH600519'], start_time='20200101', end_time='20201231')
the result is :
close
high
datetime
instrument
2020-01-02
SH600519
237.950500
241.121780
2020-01-03
SH600519
227.118515
235.213028
2020-01-06
SH600519
226.998474
230.138153
2020-01-07
SH600519
230.481384
231.422653
2020-01-08
SH600519
229.135818
229.135818
...
...
...
...
2020-09-21
SH600519
355.087494
360.617889
2020-09-22
SH600519
354.850983
359.386536
2020-09-23
SH600519
351.504211
354.919189
2020-09-24
SH600519
347.102905
353.400238
2020-09-25
SH600519
348.336365
351.508484
180 rows × 2 columns
but the post-adjustd price data from choice database is:
Yes, I know the price in qlib and the actual price they are pertty similar: I only need to apply a value resize to make the equal. But if I update the qlib data with my own csv actual backward adjusted data, will qlib remember to do the reverse resize to keep the price series consistent?
Yes, I read the docs about adjusted price data and how to dump my own csv to bins. But the problem is, the price I got in qlib is far diferent from actual price:
for stock SH600519, use code below: qdl = QlibDataLoader(config=(['$close', '$high'],['close', 'high'])) qdl.load(instruments=['SH600519'], start_time='20200101', end_time='20201231') the result is :
180 rows × 2 columns
but the post-adjustd price data from choice database is:![image](https://github.com/microsoft/qlib/assets/5120393/98720448-dcef-4a0f-80e9-b46b14107b56)
Yes, I know the price in qlib and the actual price they are pertty similar: I only need to apply a value resize to make the equal. But if I update the qlib data with my own csv actual backward adjusted data, will qlib remember to do the reverse resize to keep the price series consistent?