Deterministic or stochastic trend: decision on the basis of the augmented Dickey-Fuller test
Time series with deterministic and stochastic trends possess different memory characteristics and exhibit dissimilar long-range development. Trending series are nonstationary and must be transformed to be stabilized. The choice of correct transformation depends on patterns of nonstationarity in the...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article (Journal) |
| Language: | English |
| Published: |
March 24, 2010
|
| In: |
Methodology
Year: 2010, Volume: 6, Issue: 2, Pages: 83-92 |
| ISSN: | 1614-2241 |
| DOI: | 10.1027/1614-2241/a000009 |
| Online Access: | Verlag, lizenzpflichtig, Volltext: https://doi.org/10.1027/1614-2241/a000009 Verlag, lizenzpflichtig, Volltext: https://econtent.hogrefe.com/doi/10.1027/1614-2241/a000009 
|
| Author Notes: | Tetiana Stadnytska |
| Summary: | Time series with deterministic and stochastic trends possess different memory characteristics and exhibit dissimilar long-range development. Trending series are nonstationary and must be transformed to be stabilized. The choice of correct transformation depends on patterns of nonstationarity in the data. Inappropriate transformations are consequential for subsequent analysis and should be omitted. The objectives of this article are (1) to introduce unit root testing procedures, (2) to evaluate the strategies for distinguishing between stochastic and deterministic alternatives by means of Monte Carlo experiments, and (3) to demonstrate their implementation on empirical examples using SAS for Windows. |
|---|---|
| Item Description: | Gesehen am 12.10.2023 |
| Physical Description: | Online Resource |
| ISSN: | 1614-2241 |
| DOI: | 10.1027/1614-2241/a000009 |