Modeling Seasonal ARMA Processes.
Abstract
Gray, Kelley, and McIntire (1978) have introduced a method, based on arrays of numbers called R- and S-arrays, for identifying p and q in an ARMA (p,q) process. In addition, they have illustrated how the same method is useful in detecting nonstationary factors in an observed process, and in suggesting an appropriate transformation to stationarity. In the present paper special attention is given to the problem of modeling seasonal ARMA processes using the S-array method. A general definition is given for a seasonal process, and the procedure for identifying and modeling such processes is discussed in detail. Additionally, an interesting theorem characterizing the S-arrays (based upon the sample autocorrelation) of seasonal processes is stated and a proof indicated. Finally, a data set (the international airline data) which exhibits the properties of a seasonal process is analyzed using the method discussed, and two models for the data are proposed. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 1980
- Accession Number
- ADA091515
Entities
People
- H. L. Gray
- Jeffrey D. Hart
Organizations
- Southern Methodist University