Trailing-Edge
-
PDP-10 Archives
-
decus_20tap2_198111
-
decus/20-0026/smirn.ssp
There are 2 other files named smirn.ssp in the archive. Click here to see a list.
C SMIR 10
C ..................................................................SMIR 20
C SMIR 30
C SUBROUTINE SMIRN SMIR 40
C SMIR 50
C PURPOSE SMIR 60
C COMPUTES VALUES OF THE LIMITING DISTRIBUTION FUNCTION FOR SMIR 70
C THE KOLMOGOROV-SMIRNOV STATISTIC. SMIR 80
C SMIR 90
C USAGE SMIR 100
C CALL SMIRN(X,Y) SMIR 110
C SMIR 120
C DESCRIPTION OF PARAMETERS SMIR 130
C X - THE ARGUMENT OF THE SMIRN FUNCTION SMIR 140
C Y - THE RESULTANT SMIRN FUNCTION VALUE SMIR 150
C SMIR 160
C REMARKS SMIR 170
C Y IS SET TO ZERO IF X IS NOT GREATER THAN 0.27, AND IS SET SMIR 180
C TO ONE IF X IS NOT LESS THAN 3.1. ACCURACY TESTS WERE MADE SMIR 190
C REFERRING TO THE TABLE GIVEN IN THE REFERENCE BELOW. SMIR 200
C TWO ARGUMENTS, X= 0.62, AND X = 1.87 GAVE RESULTS WHICH SMIR 210
C DIFFER FROM THE SMIRNOV TABLES BY 2.9 AND 1.9 IN THE 5TH SMIR 220
C DECIMAL PLACE. ALL OTHER RESULTS SHOWED SMALLER ERRORS, SMIR 230
C AND ERROR SPECIFICATIONS ARE GIVEN IN THE ACCURACY TABLES SMIR 240
C IN THIS MANUAL. IN DOUBLE PRECISION MODE, THESE SAME SMIR 250
C ARGUMENTS RESULTED IN DIFFERENCES FROM TABLED VALUES BY 3 SMIR 260
C AND 2 IN THE 5TH DECIMAL PLACE. IT IS NOTED IN SMIR 270
C LINDGREN (REFERENCE BELOW) THAT FOR HIGH SIGNIFICANCE LEVELSSMIR 280
C (SAY, .01 AND .05) ASYMPTOTIC FORMULAS GIVE VALUES WHICH ARESMIR 290
C TOO HIGH ( BY 1.5 PER CENT WHEN N = 80). THAT IS, AT HIGH SMIR 300
C SIGNIFICANCE LEVELS, THE HYPOTHESIS OF NO DIFFERENCE WILL BESMIR 310
C REJECTED TOO SELDOM USING ASYMPTOTIC FORMULAS. SMIR 320
C SMIR 330
C SUBROUTINES AND FUNCTION SUBPROGRAMS REQUIRED SMIR 340
C NONE SMIR 350
C SMIR 360
C METHOD SMIR 370
C THE METHOD IS DESCRIBED BY W. FELLER-ON THE KOLMOGOROV- SMIR 380
C SMIRNOV LIMIT THEOREMS FOR EMPIRICAL DISTRIBUTIONS- ANNALS SMIR 390
C OF MATH. STAT., 19, 1948, 177-189, BY N. SMIRNOV--TABLE SMIR 400
C FOR ESTIMATING THE GOODNESS OF FIT OF EMPIRICAL SMIR 410
C DISTRIBUTIONS- ANNALS OF MATH. STAT., 19, 1948, 279-281, SMIR 420
C AND GIVEN IN LINDGREN, STATISTICAL THEORY, THE MACMILLAN SMIR 430
C COMPANY, N. Y., 1962. SMIR 440
C SMIR 450
C ..................................................................SMIR 460
C SMIR 470
SUBROUTINE SMIRN(X,Y) SMIR 480
C DOUBLE PRECISION X,Q1,Q2,Q4,Q8,Y SMIR 490
C SMIR 500
C IF A DOUBLE PRECISION VERSION OF THIS ROUTINE IS DESIRED, THE CSMIR 510
C IN COLUMN ONE OF THE DOUBLE PRECISION CARD ABOVE SHOULD BE SMIR 520
C REMOVED, AND THE C IN COLUMN ONE OF THE STATEMENTS NUMBERED SMIR 530
C C 3, C 5, AND C 8 SHOULD BE REMOVED, AND THESE CARDS SMIR 540
C SHOULD REPLACE THE STATEMENTS NUMBERED 3, 5, AND 8, SMIR 550
C RESPECTIVELY. ALL ROUTINES CALLING THIS ROUTINE MUST ALSO SMIR 560
C PROVIDE DOUBLE PRECISION ARGUMENTS TO THIS ROUTINE. SMIR 570
C SMIR 580
C ..................................................................SMIR 590
C SMIR 600
IF(X-.27)1,1,2 SMIR 610
1 Y=0.0 SMIR 620
GO TO 9 SMIR 630
2 IF(X-1.0)3,6,6 SMIR 640
3 Q1=EXP(-1.233701/X**2) SMIR 650
C 3 Q1=DEXP(-1.233700550136170/X**2) SMIR 660
Q2=Q1*Q1 SMIR 670
Q4=Q2*Q2 SMIR 680
IF(Q4 .LT. 1E-15) Q4=0
Q8=Q4*Q4 SMIR 690
IF(Q8-1.0E-15)4,5,5 SMIR 700
4 Q8=0.0 SMIR 710
5 Y=(2.506628/X)*Q1*(1.0+Q8*(1.0+Q8*Q8)) SMIR 720
C 5 Y=(2.506628274631001/X)*Q1*(1.0D0+Q8*(1.0D0+Q8*Q8)) SMIR 730
GO TO 9 SMIR 740
6 IF(X-3.1)8,7,7 SMIR 750
7 Y=1.0 SMIR 760
GO TO 9 SMIR 770
8 Q1=EXP(-2.0*X*X) SMIR 780
C 8 Q1=DEXP(-2.0D0*X*X) SMIR 790
Q2=Q1*Q1 SMIR 800
Q4=Q2*Q2 SMIR 810
Q8=Q4*Q4 SMIR 820
Y=1.0-2.0*(Q1-Q4+Q8*(Q1-Q8)) SMIR 830
9 RETURN SMIR 840
END SMIR 850