3-Month US Treasury Bills

该数据集包含了500个交易日的3个月期美国国债数据，用于训练和评估所提出的神经随机微分方程模型。该数据集在神经随机微分方程的变分推断背景下使用，并与非线性模型和线性模型进行了比较。数据集的规模属于金融数据范畴，任务则是利用神经随机微分方程进行时间序列建模。

This dataset contains 3-month U.S. Treasury bill data covering 500 trading days, and is used for training and evaluating the proposed neural stochastic differential equation model. It is applied under the variational inference framework for neural stochastic differential equations, and compared with both nonlinear and linear benchmark models. The dataset falls within the scope of financial data, with its associated task being time series modeling using neural stochastic differential equations.