Please use this identifier to cite or link to this item: http://scholarbank.nus.edu.sg/handle/10635/146988
Title: SMALL AND MEDIUM BAYESIAN VECTOR AUTOREGRESSIONS: FORECASTING GROWTH DOMESTIC PRODUCT & INFLATION IN HONG KONG.
Authors: MICHELLE WANG SHUTING
Keywords: Bayesian vector autoregressive, forecasting
Issue Date: 9-Apr-2018
Citation: MICHELLE WANG SHUTING (2018-04-09). SMALL AND MEDIUM BAYESIAN VECTOR AUTOREGRESSIONS: FORECASTING GROWTH DOMESTIC PRODUCT & INFLATION IN HONG KONG.. ScholarBank@NUS Repository.
Abstract: When handling large datasets, vector autoregressive (VAR) models often encounter over-parameterization and over-fitting. To overcome these shortcomings, Banbura et al. (2010) developed a method of parameter shrinkage in Bayesian vector autoregressive (BVAR) models that ensure that the in-sample fit of small, medium, and large BVARs are equal. This paper capitalizes on Banbura et al.’s (2010) developments and applies BVAR models with Bayesian shrinkage to forecasting Hong Kong’s output and inflation from 1994:Q1 to 2017:Q2. We compare a 2-variable Small BVAR, a medium Domestic 1 BVAR with 10 domestic variables, a small Domestic 2 BVAR with 4 domestic variables, and a medium International BVAR with 20 domestic and international variables to VAR specifications. The results show that while the BVAR models do not outperform the VAR models in root mean-squared forecast error (RMSFE) values, they are equally competitive and if not outperform them in forecasting turning points, especially for inflation.
URI: http://scholarbank.nus.edu.sg/handle/10635/146988
Appears in Collections:Bachelor's Theses (Restricted)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
Michelle Wang Shuting AY1718 Sem 2.pdf664.68 kBAdobe PDF

RESTRICTED

NoneLog In

Page view(s)

17
checked on Sep 20, 2018

Download(s)

3
checked on Sep 20, 2018

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.