Abstract: In this paper a new algorithm to identify Auto-Regressive Exogenous Models (ARX) based on Twin Support Vector Machine Regression (TSVR) has been developed. The model is determined by minimizing two ε insensitive loss functions. One of them determines the ε1-insensitive down bound regressor while the other determines the ε2-insensitive up-bound regressor. The algorithm is compared to S...

To forecast the short-term wind power precisely, this paper proposes a hybrid strategy which consists of a nonlinear dimensionality reduction component by auto-encoder network and a forecasting component based on Sparse Bayesian Regression optimized by Artificial Bee Colony Optimization. The proposed model can predict wind power curve per hour with a lead time of 3hours. Finally, an experiment ...

The amount of credit in the economy is a heterogeneous aggregate that can be analyzed across different dimensions. Considering such dimensions provides insights into effect monetary policy interventions because components are observed to respond differently. Several possible motivations behind differential response and those relate either demand supply factors intrinsic transmission mechanism p...

This paper discusses three modelling techniques, which apply to multiple time series data that correspond to different spatial locations (spatial time series). The first two methods, namely the Space-Time ARIMA (STARIMA) and the Bayesian Vector Autoregressive (BVAR) model with spatial priors apply when interest lies on the spatio-temporal evolution of a single variable. The former is better sui...

In this paper, we develop novel Markov chain Monte Carlo sampling methodology for Bayesian Cointegrated Vector Auto Regression (CVAR) models. Here we focus on two novel extensions to the sampling methodology for the CVAR posterior distribution. The first extension we develop replaces the popular sampling methodology of the griddy Gibbs sampler with an automated alternative which is based on an ...

This Paper describes a procedure for constructing theory restricted prior distributions for BVAR models. The Bayes Factor, which is obtained without any additional computational effort, can be used to assess the plausibility of the restrictions imposed on the VAR parameter vector by competing DSGE models. In other words, it is possible to rank the amount of abstraction implied by each DSGE mode...

Nowcasting can play a key role in giving policymakers timelier insight to data published with significant time lag, such as final GDP figures. Currently, there are plethora of methodologies and approaches for practitioners choose from. However, lacks comprehensive comparison these disparate terms predictive performance characteristics. This paper addresses that deficiency by examining the 12 di...

We examine the real-time forecasting performance of Bayesian VARs (BVARs) of different sizes using an unbalanced data panel. In a real-time out-of-sample forecasting exercise, we find that our BVAR methodology outperforms univariate and VAR benchmarks, and produces comparable forecast accuracy to the judgementally-adjusted forecasts produced internally at the Reserve Bank of New Zealand. We ana...