Forecasting with a noncausal VAR model

Henri Nyberg, Pentti Saikkonen

Forskningsoutput: ArbetsdokumentDiskussionsartiklarVetenskaplig

Sammanfattning

We propose simulation-based forecasting methods for the noncausal vector autoregressive model proposed by Lanne and Saikkonen (2012). Simulation or numerical methods are required because the prediction problem is generally nonlinear and, therefore, its analytical solution is not available. It turns out that different special cases of the model call for different simulation procedures. Simulation experiments demonstrate that gains in forecasting accuracy are achieved by using the correct noncausal VAR model instead of its conventional causal counterpart. In an empirical application, a noncausal VAR model comprised of U.S. inflation and marginal cost turns out superior to the best-fitting conventional causal VAR model in forecasting inflation.
We propose simulation-based forecasting methods for the noncausal vector autoregressive model proposed by Lanne and Saikkonen (2012). Simulation or numerical methods are required because the prediction problem is generally nonlinear and, therefore, its analytical solution is not available. It turns out that different special cases of the model call for different simulation procedures. Simulation experiments demonstrate that gains in forecasting accuracy are achieved by using the correct noncausal VAR model instead of its conventional causal counterpart. In an empirical application, a noncausal VAR model comprised of U.S. inflation and marginal cost turns out superior to the best-fitting conventional causal VAR model in forecasting inflation.
Originalspråkengelska
UtgivningsortHelsinki
FörlagBank of Finland
Antal sidor37
ISBN (elektroniskt)978-952-462-828-0
StatusPublicerad - nov 2012
MoE-publikationstypD4 Publicerad utvecklings- eller forskningsrapport eller studie

Vetenskapsgrenar

  • 112 Statistik
  • 511 Nationalekonomi

Citera det här

Nyberg, H., & Saikkonen, P. (2012). Forecasting with a noncausal VAR model. (Bank of Finland Research Discussion Papers; Nr. 33/2012). Helsinki: Bank of Finland.
Nyberg, Henri ; Saikkonen, Pentti. / Forecasting with a noncausal VAR model. Helsinki : Bank of Finland, 2012. (Bank of Finland Research Discussion Papers; 33/2012).
@techreport{c23f6066c8554a0ebfbdb64171cea3a7,
title = "Forecasting with a noncausal VAR model",
abstract = "We propose simulation-based forecasting methods for the noncausal vector autoregressive model proposed by Lanne and Saikkonen (2012). Simulation or numerical methods are required because the prediction problem is generally nonlinear and, therefore, its analytical solution is not available. It turns out that different special cases of the model call for different simulation procedures. Simulation experiments demonstrate that gains in forecasting accuracy are achieved by using the correct noncausal VAR model instead of its conventional causal counterpart. In an empirical application, a noncausal VAR model comprised of U.S. inflation and marginal cost turns out superior to the best-fitting conventional causal VAR model in forecasting inflation.",
keywords = "112 Statistics and probability, 511 Economics",
author = "Henri Nyberg and Pentti Saikkonen",
year = "2012",
month = "11",
language = "English",
series = "Bank of Finland Research Discussion Papers",
publisher = "Bank of Finland",
number = "33/2012",
address = "Finland",
type = "WorkingPaper",
institution = "Bank of Finland",

}

Nyberg, H & Saikkonen, P 2012 'Forecasting with a noncausal VAR model' Bank of Finland Research Discussion Papers, nr. 33/2012, Bank of Finland, Helsinki.

Forecasting with a noncausal VAR model. / Nyberg, Henri; Saikkonen, Pentti.

Helsinki : Bank of Finland, 2012. (Bank of Finland Research Discussion Papers; Nr. 33/2012).

Forskningsoutput: ArbetsdokumentDiskussionsartiklarVetenskaplig

TY - UNPB

T1 - Forecasting with a noncausal VAR model

AU - Nyberg, Henri

AU - Saikkonen, Pentti

PY - 2012/11

Y1 - 2012/11

N2 - We propose simulation-based forecasting methods for the noncausal vector autoregressive model proposed by Lanne and Saikkonen (2012). Simulation or numerical methods are required because the prediction problem is generally nonlinear and, therefore, its analytical solution is not available. It turns out that different special cases of the model call for different simulation procedures. Simulation experiments demonstrate that gains in forecasting accuracy are achieved by using the correct noncausal VAR model instead of its conventional causal counterpart. In an empirical application, a noncausal VAR model comprised of U.S. inflation and marginal cost turns out superior to the best-fitting conventional causal VAR model in forecasting inflation.

AB - We propose simulation-based forecasting methods for the noncausal vector autoregressive model proposed by Lanne and Saikkonen (2012). Simulation or numerical methods are required because the prediction problem is generally nonlinear and, therefore, its analytical solution is not available. It turns out that different special cases of the model call for different simulation procedures. Simulation experiments demonstrate that gains in forecasting accuracy are achieved by using the correct noncausal VAR model instead of its conventional causal counterpart. In an empirical application, a noncausal VAR model comprised of U.S. inflation and marginal cost turns out superior to the best-fitting conventional causal VAR model in forecasting inflation.

KW - 112 Statistics and probability

KW - 511 Economics

M3 - Discussion paper

T3 - Bank of Finland Research Discussion Papers

BT - Forecasting with a noncausal VAR model

PB - Bank of Finland

CY - Helsinki

ER -

Nyberg H, Saikkonen P. Forecasting with a noncausal VAR model. Helsinki: Bank of Finland. 2012 nov. (Bank of Finland Research Discussion Papers; 33/2012).