Interpolation of economic time series, with application to German and Swedish data (Q4356800)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 1070336
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Interpolation of economic time series, with application to German and Swedish data |
scientific article; zbMATH DE number 1070336 |
Statements
5 October 1997
0 references
interpolation of economic time series
0 references
Chow-Lin procedure
0 references
interpolations
0 references
Bayesian procedure
0 references
0.792671263217926
0 references
0.7559306025505066
0 references
0.7548908591270447
0 references
0.7431878447532654
0 references
0.7380148768424988
0 references
Interpolation of economic time series, with application to German and Swedish data (English)
0 references
Authors' conclusion: We have found that the standard Chow-Lin procedure provides good interpolations for a number of German data sets covering reasonably long time spans, and appears to be largely free of the so-called discontinuity problem according to which jumps may be expected to be observed from one low-frequency period to the next. Its main disadvantage, that it permits negative interpolated values, has been surmounted by implementation of a quadratic-programming algorithm which need only be applied separately to each low-frequency period for which negative interpolated values are obtained. This procedure is found to provide good interpolations in a simulated data set, and has been applied to German and Swedish investment data. Finally, for short periods, a Bayesian procedure using prior information from an earlier data set performs better than standard procedures using information from the entire data set. This suggests that for countries that have only short and low-frequency time series, use of prior information and possibly also high-frequency related series from other countries could be expected to provide fairly reliable interpolated high-frequency series. For the entire collection see [Zbl 0844.00014].
0 references