Description PROGRAM LEVEL2: compiled with Compaq Visual Fortran 6.6 (2006) PUBLICATION CRT-972: LEVEL: The L-2.0 program for BC-DAUHESEQ regressions --- Box-Cox Direc AUtoregressive HEteroskedastic Single Equation models. by TRAN LIEM, MARC GAUDRY and ULRICH BLUM. Universite de Montreal, Centre de recherche sur les transports, June 1998. The main program contains two minimization algorithms : 1- r.fletcher and m.j.d.powell,"a rapidly convergent descent method for minimization", computer journal,6,pp.163,168 (1963), 2- m.j.d.powell,"an efficient method for finding the minimum of a function of several variables without calculating derivatives", computer journal,7,pp.155,162 (1964).
Source file: level2.f
I/O Operations: Unit ID Unit No Access Form Operation 6 SEQ FMTD W Operation codes A=rewind,B=backspace,C=close,E=endfile I=inquire,O=open,R=read,W=write
Intrinsic Functions Called DOUBLE PRECISION DABS, DBLE, DSQRT REAL SNGL
External Functions and Subroutines Called DOUBLE PRECISION VP SUBROUTINE FCN
Parameter Variables Used INTEGER N (N = 35) INTEGER N1 (N1 = N+1)
Local Variables (+ indicates altered content) DOUBLE PRECISION +A, +AL, +AM, +AME DOUBLE PRECISION +AMIN, +AST, +AV, +AVR DOUBLE PRECISION +B, +BE, +C, +D(N) DOUBLE PRECISION +DA, +DB, +DEPSI, +DF DOUBLE PRECISION +DFG, +DG, +DIR, +DIRIN(N1) DOUBLE PRECISION +DP, +DWGHT, +E(2), +EPSILON DOUBLE PRECISION +ET(N,N), +F, +FA, +G(N) DOUBLE PRECISION +GAM, +P(N), +S(N1), +SA DOUBLE PRECISION +SP, +SS, +ST, +STEP DOUBLE PRECISION +STJ, +WT(N), +X(N), +X1 DOUBLE PRECISION +X2, +XN, +XS(N1,N), +Y(N) DOUBLE PRECISION +Y1, +Y2 INTEGER +I, +ISW(6), +J, +JM INTEGER +K, +L, +M, +NE INTEGER +NEQ, +NFCN, +NOR, +NP INTEGER +NP1, NPAR, +NPFN, +NS INTEGER +NST LOGICAL +DERCAL
Referenced Common Block Variables (+ indicates altered content) %BLANK INTEGER NLIST, +NOPRINT %BLANK INTEGER NSTEP %BLANK REAL EPSI, WGHT