What are the trade-offs between using a random forest model versus a gradient-boosted decision tree for predicting short-term asset price movements?
