Enhanced Bayesian Neural Networks for Macroeconomics and Finance

by   Niko Hauzenberger, et al.
Università Bocconi

We develop Bayesian neural networks (BNNs) that permit to model generic nonlinearities and time variation for (possibly large sets of) macroeconomic and financial variables. From a methodological point of view, we allow for a general specification of networks that can be applied to either dense or sparse datasets, and combines various activation functions, a possibly very large number of neurons, and stochastic volatility (SV) for the error term. From a computational point of view, we develop fast and efficient estimation algorithms for the general BNNs we introduce. From an empirical point of view, we show both with simulated data and with a set of common macro and financial applications that our BNNs can be of practical use, particularly so for observations in the tails of the cross-sectional or time series distributions of the target variables.


page 25

page 27


Modeling Multiple Irregularly Spaced Financial Time Series

In this paper we propose univariate volatility models for irregularly sp...

Event-Based Dynamic Banking Network Exploration for Economic Anomaly Detection

The instability of financial system issues might trigger a bank failure,...

A maximum entropy network reconstruction of macroeconomic models

In this article the problem of reconstructing the pattern of connection ...

A Bayesian Semiparametric Vector Multiplicative Error Model

Interactions among multiple time series of positive random variables are...

Toward Reliable Neural Specifications

Having reliable specifications is an unavoidable challenge in achieving ...

Please sign up or login with your details

Forgot password? Click here to reset