search

tomespel / kairos-volatility-prediction

Advanced machine learning techniques for volatility prediction on financial markets. The Kairos package was developed for CFM’s ENS DataChallenge 2018.
https://challengedata.ens.fr/en/challenge/34/volatility_prediction_in_financial_markets.html
Apache License 2.0
0 stars 0 forks source link

Raw data preprocessing and class KairosDay #4

Closed mehditomas closed 6 years ago

mehditomas commented 6 years ago

Dealing with raw data

The notebooks were implemented with data in its original form, which was easy to manipulate. It might not be necessary to reformat the time stamp format of the different assets given that. It might add more complexity and force us to modify code.

What do you think?

tomespel commented 6 years ago

The current time stamps format isn't actually time stamps, there are strings. Besides, it is always the same time base used for both volatility and returns. In the current class version, there has been no addition or change of the time stamp, it has just been removed to ease the computations and the output size. Recall what matters is the sequence - and subsequently, the position in the sequence - of the piece of data, instead of the actual time, that is used for computation. Does that work for you?

tomespel commented 6 years ago

Issue continued in issue #6 and closed.