TabNet Forecasting of Short-Term Silver Futures Direction

Summary

This study predicts short-term directional changes in a futures contract using features drawn from technical analysis, order flow, and order-book data. It trains a TabNet deep learning model on silver futures listed on the Shanghai Futures Exchange, combining these inputs to classify the direction of the next movement.

The reported accuracy is 0.601 for the selected evaluation period. The document provides little detail about the period, data preparation, validation design, trading costs, or performance relative to a baseline. The result therefore describes a prediction experiment rather than establishing that the model would produce profitable trades or generalize to other contracts and market conditions.

Key ideas

• The model combines technical indicators, order-flow features, and order-book data.
• TabNet is used to predict short-term directional changes in silver futures.
• The reported accuracy is 0.601 on the selected period.
• The brief description does not establish profitability or performance outside the studied contract and period.

Tags

# Neural Network and Order Flow, Technical Analysis: Predicting short-term direction of futures contract


# Neural Network and Order Flow, Technical Analysis: Predicting short-term direction of futures contract









Predictions of short-term directional movement of the futures contract can be challenging as its pricing is often based on multiple complex dynamic conditions. This work presents a method for predicting the short-term directional movement of an underlying futures contract. We engineered a set of features from technical analysis, order flow, and order-book data. Then, Tabnet, a deep learning neural network, is trained using these features. We train our model on the Silver Futures Contract listed on Shanghai Futures Exchange and achieve an accuracy of 0.601 on predicting the directional change during the selected period.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.