Program g02dafe
! G02DAF Example Program Text
! Mark 27.1 Release. NAG Copyright 2020.
! .. Use Statements ..
Use nag_library, Only: g02buf, g02daf, nag_wp
! .. Implicit None Statement ..
Implicit None
! .. Parameters ..
Integer, Parameter :: nin = 5, nout = 6
! .. Local Scalars ..
Real (Kind=nag_wp) :: aic, arsq, en, mult, rsq, rss, sw, &
tol
Integer :: i, idf, ifail, ip, irank, ldq, ldx, &
lwt, m, n
Logical :: svd
Character (1) :: mean, weight
! .. Local Arrays ..
Real (Kind=nag_wp), Allocatable :: b(:), cov(:), h(:), p(:), q(:,:), &
res(:), se(:), wk(:), wt(:), x(:,:), &
y(:)
Real (Kind=nag_wp) :: c(1), wmean(1)
Integer, Allocatable :: isx(:)
! .. Intrinsic Procedures ..
Intrinsic :: count, log, real
! .. Executable Statements ..
Write (nout,*) 'G02DAF Example Program Results'
Write (nout,*)
! Skip heading in data file
Read (nin,*)
! Read in the problem size
Read (nin,*) n, m, weight, mean
If (weight=='W' .Or. weight=='w') Then
lwt = n
Else
lwt = 0
End If
ldx = n
Allocate (x(ldx,m),y(n),wt(lwt),isx(m))
! Read in data
If (lwt>0) Then
Read (nin,*)(x(i,1:m),y(i),wt(i),i=1,n)
Else
Read (nin,*)(x(i,1:m),y(i),i=1,n)
End If
! Read in variable inclusion flags
Read (nin,*) isx(1:m)
! Calculate IP
ip = count(isx(1:m)>0)
If (mean=='M' .Or. mean=='m') Then
ip = ip + 1
End If
ldq = n
Allocate (b(ip),cov((ip*ip+ip)/2),h(n),p(ip*(ip+ &
2)),q(ldq,ip+1),res(n),se(ip),wk(ip*ip+5*(ip-1)))
! Use suggested value for tolerance
tol = 0.000001E0_nag_wp
! Fit general linear regression model
ifail = -1
Call g02daf(mean,weight,n,x,ldx,m,isx,ip,y,wt,rss,idf,b,se,cov,res,h,q, &
ldq,svd,irank,p,tol,wk,ifail)
If (ifail/=0) Then
If (ifail/=5) Then
Go To 100
End If
End If
! Calculate (weighted) total sums of squares, adjusted for mean if
! required
! If in G02DAF, an intercept is added to the regression by including a
! column of 1's in X, rather than by using the MEAN argument then
! MEAN = 'M' should be used in this call to G02BUF.
ifail = 0
Call g02buf(mean,weight,n,1,y,n,wt,sw,wmean,c,ifail)
! Get effective number of observations (=N if there are no zero weights)
en = real(idf+irank,kind=nag_wp)
! Calculate R-squared, corrected R-squared and AIC
rsq = 1.0_nag_wp - rss/c(1)
If (mean=='M' .Or. mean=='m') Then
mult = (en-1.0E0_nag_wp)/(en-real(irank,kind=nag_wp))
Else
mult = en/(en-real(irank,kind=nag_wp))
End If
arsq = 1.0_nag_wp - mult*(1.0_nag_wp-rsq)
aic = en*log(rss/en) + 2.0_nag_wp*real(irank,kind=nag_wp)
! Display results
If (svd) Then
Write (nout,99999) 'Model not of full rank, rank = ', irank
Write (nout,*)
End If
Write (nout,99998) 'Residual sum of squares = ', rss
Write (nout,99999) 'Degrees of freedom = ', idf
Write (nout,99998) 'R-squared = ', rsq
Write (nout,99998) 'Adjusted R-squared = ', arsq
Write (nout,99998) 'AIC = ', aic
Write (nout,*)
Write (nout,*) 'Variable Parameter estimate Standard error'
Write (nout,*)
If (ifail==0) Then
Write (nout,99997)(i,b(i),se(i),i=1,ip)
Else
Write (nout,99996)(i,b(i),i=1,ip)
End If
Write (nout,*)
Write (nout,*) ' Obs Residuals H'
Write (nout,*)
Write (nout,99997)(i,res(i),h(i),i=1,n)
100 Continue
99999 Format (1X,A,I4)
99998 Format (1X,A,E12.4)
99997 Format (1X,I6,2E20.4)
99996 Format (1X,I6,E20.4)
End Program g02dafe