Journal of Computational Finance

Risk.net

Nowcasting networks

Marc Chataigner, Stéphane Crépey and Jiang Pu

  • A neural network-based compression/completion methodology for financial nowcasting, ie, completion of gridded values, interpolation or outlier detection, in the context of financial time series of curves, surfaces, etc
  • An original network architecture amenable to the treatment of data defined on variable grids (by far the most common situation in financial nowcasting applications), so that PCA or standard autoencoders are not applicable
  • Three case studies on real data sets: repo curves, equity derivative surfaces and at-the-money swaption surfaces
  • Github with code and data

We devise a neural network-based compression/completion methodology for financial nowcasting. The latter is meant in a broad sense, encompassing completion of gridded values, interpolation and outlier detection, in the context of financial time series of curves or surfaces. (It is also applicable in higher dimensions, at least in theory.) In particular, we introduce an original architecture amenable to the treatment of data defined at variable grid nodes (by far the most common situation in financial nowcasting applications, where principal component analysis (PCA) and classical autoencoder methods are not applicable). This is illustrated by three case studies on real data sets. First, we introduce our approach on repurchase agreement curves data (with a moving time-to-maturity as calendar time passes). Second, we show that our approach outperforms elementary interpolation benchmarks on an equity derivative surfaces data set (again, with a moving time-to-maturity). We also obtain a satisfying performance for outlier detection and surface completion. Third, we benchmark our approach against PCA on at-the-money swaption surfaces redefined at constant expiry/tenor grid nodes. Our approach is then shown to perform as well as (even if not obviously better than) the PCA (which, however, is not applicable to the native, raw data defined on a moving time-to-expiry grid).

To continue reading...

You need to sign in to use this feature. If you don’t have a Risk.net account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here: