We present in this article a non-parametric value-at-risk (VaR+CVaR) algorithm that remains accurate for an arbitrarily large number of underlying positions. The algorithm solves the two inherent problems of VaR estimation. First, past history is not directly applicable to the future, but all predictions of the future are based on the past. Second, VaR estimation is equivalent to modeling a single corner of a high-dimensional space (the corner where all bets lose simultaneously). The algorithm only uses mathematical methods that strictly do not degrade in accuracy at high-dimensions. Historical data are then directly incorporated with all high-dimensional relationships present, without manipulation. We test the algorithm with an ensemble of 500 portfolios with random positions across 49 distinct liquid futures of different expiries (VIX, equity indexes, gov. bonds, rates, energy, metals, livestock, agriculture, and softs). All VaR estimations are performed strictly blind to the future. The median portfolio rate of loss exceeding the 99
Accurate stock trend forecasting is a central challenge in financial economics due to the highly nonlinear and interdependent nature of market dynamics. Traditional statistical and machine learning models often fall short in modeling complex temporal dependencies and interstock relationships. This paper proposes a novel framework based on graph neural networks for short-term stock trend prediction. By representing sliding windows of historical stock prices as temporal graphs and incorporating domain-specific technical indicators-namely, the 5-day moving average and the 14-day relative strength index-the proposed model captures both local temporal interactions and indicator-informed momentum patterns. Experimental results highlight the potential of graph-based modeling in capturing short-term trading signals and improving the predictive performance of financial forecasting systems.
