1. CatBoost based Jane Street Market Forecast Model
- Author
-
Yubing He, Kikko, Maoyuan Li, and Ouyang Jingze
- Subjects
Feature engineering ,Stock market prediction ,Profit (accounting) ,Boosting (machine learning) ,Artificial neural network ,business.industry ,Computer science ,Machine learning ,computer.software_genre ,Task (project management) ,Value (economics) ,Feature (machine learning) ,Artificial intelligence ,business ,computer - Abstract
Nowadays, stock market prediction and trading has attracted many investors who want to make a higher profit. And a lot of researchers have paid attention on it because it is a challenging task due to the high complexity of the market. More investors put their effort to the development of a systematic approach. Many machine learning algorithms have been utilized for the prediction of action. In this paper, we adopted a CatBoost method which is a kind of boosting method leading to an optimal performance. In the feature engineering, we use the mean of other values about one kind of feature to fill NaN value. And we show the figure about the missing value distribution of the feature. We train our model on the dataset from Jane Street Market provided by kaggle website. The experiments show that our method achieves superior performance over the other machine learning approaches. Our model's unity score is 202 and 401 higher than those of Lightgbm model algorithm and Neural Network model. respectively.
- Published
- 2021