program Simple Linear Regression
      parameter (N=31)
      dimension y(N),x(N)

c   An example :
      data x/19.,25.,22.,-1.,4.,14.,21.,22.,23.,27.,
     * 29.,25.,29.,15.,29.,24.,0.,2.,26.,17.,
     *19.,9.,20.,-6.,-13.,-13.,-11.,-4.,-4.,11.,23./
      data y/28.,28.,26.,19.,16.,24.,26.,24.,24.,29.,
     * 29.,27.,31.,26.,38.,23.,13.,14.,28.,19.,
     * 19.,17.,22.,2.,4.,5.,7.,8.,14.,14.,23./

! Call function to calculate regression coefficient and others
	call SLRegress(x,y,n,a,b,q,p,umax,umin,u,dyy)

! Output result on screen
	write(*,*)
	write(*,10)a,b
10	format(/1x,'a=',f10.5,'   b=',f10.5)
	write(*,100) dyy
100	format(1x,'Total sum of squares=',f10.4)
	write(*,102) p
102	format(1x,'Regression sum of squares=',f10.4)
	write(*,104) q
104	format(1x,'Residual sum of squares=',f10.4)
	f=p/(q/(n-2))
	write(*,106) f
106	format(1x,'F=',f10.4)
	write(*,108) umax,umin,u
108	format(1x,'Maximum error=',f10.4,'  Minimum error=',f10.4,'
     * Mean error=',f10.4)
	pause 'Press any key to continue...'
! Output the Analysis-of-Variance Table
	write(*,'(//)')
	write(6,200)
200	format(/1x, 'Generic Analysis of Variance Table for the simple 
     * Linear Regression')
	write(6,202)
202	format(/1x,'------------------------------------------------------
     *---------------')
	write(6,204)
204	format(/3x,'Source       df       SS           MS')
	write(6,202)
	write(6,206)	n-1,dyy
206	format(/1x,'Total       n-1=',i2,'  SST=',f8.3)
	write(6,208)	1,p,p
208   format(/1x,'Regression    K=',i2,'  SSR=',f8.3,'   MSR=SSR/K='
     *,f8.3)
	q2=q/real(n-1-1)
	write(6,209)	n-1-1,q,q2
209   format(/1x,'Residual  n-k-1=',i2,'  SSE=',f8.3,'   MSE=SSE/(n-k-1)
     *=',f8.3)
	f=p/(q/real(n-1-1))
	write(6,220) f
220	format(/1x,'                   F=MSR/MSE=',f9.4)
	write(6,202)

	end


!------------------------------------------------------------------------
! Simple Linear Regression
!    Input parameter
!         x: Predictor Variable
!         y: Predictand Variable
!         n: Seiral length
!    Output parameter
!       a,b: Regress coefficient, y=a+bx
!         q: Residual sum of squares
!         p: Regression sum of squares
!      umax: Maximum error
!      umin: Minimum error
!         u: Mean error
!       dyy: Total sum of squares
****************************
	subroutine SLRegress(x,y,n,a,b,q,p,umax,umin,u,dyy)
	  dimension x(n),y(n)

	xx=0.0
	yy=0.0
	do  10  i=1,n
	  xx=xx+x(i)/n
	  yy=yy+y(i)/n
 10	continue
	dx=0.0
	dxy=0.0
	dyy=0.0
	do  20  i=1,n
	  q=x(i)-xx
	  dx=dx+q*q
	  dxy=dxy+q*(y(i)-yy)
	  dyy=dyy+(y(i)-yy)*(y(i)-yy)
 20	continue
	b=dxy/dx
	a=yy-b*xx
	q=0.0
	u=0.0
	p=0.0
	umax=0.0
	umin=1.0d+30
	do  30  i=1,n
	  s=a+b*x(i)
	  q=q+(y(i)-s)*(y(i)-s)
	  p=p+(s-yy)*(s-yy)
	  dx=abs(y(i)-s)
	  if(dx.gt.umax) umax=dx
	  if(dx.lt.umin) umin=dx
	  u=u+dx/n
 30	continue
	s=sqrt(q/n)
	return
	end
!------------------------------------------------------------------------