7 OTHER METHODS of of

Most of the more powerful methods apply to special types of matrices. Many ofthem apply to symmetric matrices. Generally speaking, the original matrix is transformed into a simpler form that has the same eigenvalues. Iterative methods are then employed to evaluate the eigenvalues. More information on the subject can be found in Fadeev and Fadeeva (1963), Householder (1964), Wilkinson (1965), Steward (1973), Ralston and Rabinowitz (1978), and Press, Flannery, Teukolsky, and Vetterling (1989). Numerous compute programs for solving eigenproblems can be found in the IMSL (International Mathematical and Statistics Library) library and in the EISPACK program (Argonne National Laboratory). See Rice (1983) and Smith et al. (1976) for a discussion of these programs.