Selecting the Form of Combining Regressions Based on Recur sive Prediction Criteria

Liang Kou-yuan and Ryu Keunkwan
Additional contact information
Liang Kou-yuan: National Tsing Hua University, Department of Economics
Ryu Keunkwan: University of California, Department of Economics

A chapter in Modelling and Prediction Honoring Seymour Geisser, 1996, pp 122-135 from Springer

Abstract: Abstract This paper reformulates the basic Granger and Ramanathan’s (1984) combining regression framework based on post-sample predictive accuracies. Using recursive regression techniques, this paper develops an algorithm to estimate combining weights. Under the new prediction criteria, we show that Granger and Ramanathan’s (1984) preference ordering, Method C → Method A → Method B, breaks down. To overcome this lack of ordering, the paper suggests that, by using Akaike’s (1973) information or Amemiya’s (1980) prediction criterion, one can recursively select the best form of combining regressions. Empirical examples using macroeconomic forecasts of Taiwan are presented to illustrate the validity of the theoretical arguments.

Keywords: Mean Square Error; Prediction Error; ARIMA Model; Government Purchase; Error Reduction Rate (search for similar items in EconPapers)
Date: 1996
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-1-4612-2414-3_7

Ordering information: This item can be ordered from
http://www.springer.com/9781461224143

DOI: 10.1007/978-1-4612-2414-3_7

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().