Abstract
This paper investigates the effect of seasonal adjustment filters on the identification of mixed causal-noncausal autoregressive models. By means of Monte Carlo simulations, we find that standard seasonal filters induce spurious autoregressive dynamics on white noise series, a phenomenon already
documented in the literature. Using a symmetric argument, we show that those filters also generate a spurious noncausal component in the seasonally adjusted series, but preserve (although amplify) the existence of causal and noncausal relationships. This result has important implications for modelling economic time series driven by expectation relationships. We consider inflation data on the G7 countries to illustrate these results.
documented in the literature. Using a symmetric argument, we show that those filters also generate a spurious noncausal component in the seasonally adjusted series, but preserve (although amplify) the existence of causal and noncausal relationships. This result has important implications for modelling economic time series driven by expectation relationships. We consider inflation data on the G7 countries to illustrate these results.
| Original language | English |
|---|---|
| Article number | 48 |
| Pages (from-to) | 1-22 |
| Number of pages | 22 |
| Journal | Econometrics |
| Volume | 5 |
| Issue number | 4 |
| DOIs | |
| Publication status | Published - 31 Oct 2017 |
JEL classifications
- c22 - "Single Equation Models; Single Variables: Time-Series Models; Dynamic Quantile Regressions; Dynamic Treatment Effect Models"
- e37 - Prices, Business Fluctuations, and Cycles: Forecasting and Simulation: Models and Applications
Keywords
- Inflation
- seasonal adjustment filters
- mixed causal-noncausal models
- TIME-SERIES
- UNIT-ROOT TESTS
- inflation