Forecasting the Global Electronics Cycle with Leading Indicators: A VAR Approach
Developments in the global electronics industry are typically monitored by tracking indicators that span a whole spectrum of activities in the sector. However, these indicators invariably give mixed signals at each point in time, thereby hampering efforts at prediction. In this paper, we propose a u...
Saved in:
| Main Author: | |
|---|---|
| Format: | text |
| Language: | English |
| Published: |
Institutional Knowledge at Singapore Management University
2004
|
| Subjects: | |
| Online Access: | https://ink.library.smu.edu.sg/soe_research/830 https://ink.library.smu.edu.sg/context/soe_research/article/1829/viewcontent/Forecasting_the_Global_Electronics_Cycle_with_Leading_Indicators_.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Singapore Management University |
| Language: | English |
| Summary: | Developments in the global electronics industry are typically monitored by tracking indicators that span a whole spectrum of activities in the sector. However, these indicators invariably give mixed signals at each point in time, thereby hampering efforts at prediction. In this paper, we propose a unified framework for forecasting the global electronics cycle by constructing a VAR model that captures the economic interactions between leading indicators representing expectations, orders, inventories and prices. The ability of the indicators to presage world semiconductor sales is first demonstrated by Granger causality tests. The VAR model is then used to derive the dynamic paths of adjustment of global chip sales in response to orthogonalized shocks in each of the leading variables. These impulse response functions confirm the leading qualities of the selected indicators. Finally, out-of-sample forecasts of global chip sales are generated from a parsimonious variant of the model viz., the Bayesian VAR (BVAR), and compared with predictions from a univariate benchmark model and a bivariate model which uses a composite index of the leading indicators. An evaluation of their relative accuracy suggests that the BVAR’s forecasting performance is superior to both the univariate and composite index models. |
|---|