Paper: arXiv 2304.03877

Authors: Nikolas Michael, Mihai Cucuringu, Sam Howison

Abstract

We introduce OFTER, a time series forecasting pipeline tailored for mid-sized multivariate time series. OFTER utilizes the non-parametric models of k-nearest neighbors and Generalized Regression Neural Networks, integrated with a dimensionality reduction component. To circumvent the curse of dimensionality, we employ a weighted norm based on a modified version of the maximal correlation coefficient. The pipeline we introduce is specifically designed for online tasks, has an interpretable output, and is able to outperform several state-of-the art baselines. The computational efficacy of the algorithm, its online nature, and its ability to operate in low signal-to-noise regimes, render OFTER an ideal approach for financial multivariate time series problems, such as daily equity forecasting. Our work demonstrates that while deep learning models hold significant promise for time series forecasting, traditional methods carefully integrating mainstream tools remain very competitive alternatives with the added benefits of scalability and interpretability.

Complexity vs Empirical Score

  • Math Complexity: 5.5/10
  • Empirical Rigor: 6.0/10 (capped from 6.5: transaction costs not modelled; see penalty flags)
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper presents a solid combination of non-parametric methods with feature selection via maximal correlation coefficient targeting noisy online financial regimes. It balances theoretical adaptation of classical methods with practical empirical baseline comparisons, showing strong potential for mid-sized time series forecasting.

Research Flowchart

  flowchart TD
    A[Goal: Scalable & Interpretable Online Multivariate Time Series Forecasting] --> B[Input: Mid-Sized Financial Time Series e.g., Daily Equities]
    B --> C[Dimensionality Reduction & Feature Selection]
    C --> D[Distance Calculation via Modified Maximal Correlation Coefficient Weighted Norm]
    D --> E[Non-Parametric Modeling: k-NN & Generalized Regression Neural Networks GRNN]
    E --> F[Outperforms SOTA Baselines with High Efficacy in Low Signal-to-Noise Regimes]