A statistical physics approach to a multi-channel Wigner spiked model

In this letter we present a finite temperature approach to a high-dimensional inference problem, the Wigner spiked model, with group dependent signal-to-noise ratios. For two classes of convex and non-convex network architectures the error in the reconstruction is described in terms of the solution of a mean-field spin-glass on the Nishimori line. In the cases studied the order parameters do not fluctuate and are the solution of finite dimensional variational problems. The deep architecture is optimized in order to confine the high temperature phase where reconstruction fails.

Sharp Guarantees and Optimal Performance for Inference in Binary and Gaussian-Mixture Models

We study convex empirical risk minimization for high-dimensional inference in binary linear classification under both discriminative binary linear models, as well as generative Gaussian-mixture models. Our first result sharply predicts the statistical performance of such estimators in the proportional asymptotic regime under isotropic Gaussian features. Importantly, the predictions hold for a wide class of convex loss functions, which we exploit to prove bounds on the best achievable performance. Notably, we show that the proposed bounds are tight for popular binary models (such as signed and logistic) and for the Gaussian-mixture model by constructing appropriate loss functions that achieve it. Our numerical simulations suggest that the theory is accurate even for relatively small problem dimensions and that it enjoys a certain universality property.

Fixed‐ k inference for volatility

We present a new theory for the conduct of nonparametric inference about the latent spot volatility of a semimartingale asset price process. In contrast to existing theories based on the asymptotic notion of an increasing number of observations in local estimation blocks, our theory treats the estimation block size k as fixed. While the resulting spot volatility estimator is no longer consistent, the new theory permits the construction of asymptotically valid and easy‐to‐calculate pointwise confidence intervals for the volatility at any given point in time. Extending the theory to a high‐dimensional inference setting with a growing number of estimation blocks further permits the construction of uniform confidence bands for the volatility path. An empirically realistically calibrated simulation study underscores the practical reliability of the new inference procedures. An empirical application based on intraday data for the S&P 500 equity index reveals highly significant abrupt changes, or jumps, in the market volatility at FOMC news announcement times, validating recent uses of various high‐frequency‐based identification schemes in asset pricing finance and monetary economics.