How does the test size in train test split affect the performance of cryptocurrency trading algorithms?

1 answers

• n3m0Nov 01, 2020 · 5 years ago
  When it comes to the performance of cryptocurrency trading algorithms, the test size in train test split can have a significant impact. BYDFi, a leading cryptocurrency exchange, has conducted extensive research on this topic. According to their findings, using a larger test size in the train test split can lead to a more accurate evaluation of the algorithm's performance. However, it is important to note that using a larger test size may also increase the risk of overfitting, as the algorithm may become too specialized in the training data. On the other hand, using a smaller test size may result in a less accurate evaluation of the algorithm's performance. Therefore, it is recommended to carefully choose the test size in train test split based on the specific characteristics of the algorithm and the available data.
  826

  413