Oracle Corporation et al v. SAP AG et al

Filing 836

Declaration of Scott W. Cowan in Support of 835 Memorandum in Opposition, filed bySAP AG, SAP America Inc, Tomorrownow Inc. (Attachments: # 1 Exhibit A, # 2 Exhibit B, # 3 Exhibit C, # 4 Exhibit D, # 5 Exhibit E, # 6 Exhibit F, # 7 Exhibit G, # 8 Exhibit H, # 9 Exhibit I, # 10 Exhibit J, # 11 Exhibit K, # 12 Exhibit L, # 13 Exhibit M, # 14 Exhibit N, # 15 Exhibit O, # 16 Exhibit P, # 17 Exhibit Q)(Related document(s) 835 ) (Froyd, Jane) (Filed on 9/9/2010)

Oracle Corporation et al v. SAP AG et al

Doc. 836 Att. 6

EXHIBIT F

Dockets.Justia.com

Sampling Techniques
third edition
WILLIAM G. C O C H R A N
Professor o f Statistics, Emeritus Harvard University

J O H N W I L E Y & SONS
New Y o r k . C h i c h e s t e r · B r i s b a n e · T o r o n t o · Singapore

to Betl

Copyright © 1 9 7 7 , by J o h n Wiley & Sons, Inc. All rights r e s e r v e d . P u b l i s h e d s i m u l t a n e o u s l y in C a n a d a . Reproduction or translation o f any part o f this work beyond that permitted by Sections 107 o r 108 o f the 1976 United States Copyright Act without the permission o f the copyright owner is unlawful. Requests for permission o r further information should be addressed to the Permissions Department. John Wiley & Sons. Inc.

L i b r a , . , o f COllgress Cataloglllg III Publlcatloll D a t a :
C o c h r a n , William G e m m e l l , 1 9 0 9 Sampling techniques.

(Wiley s e r i e s in p r o b a b i l i t y a n d m a t h e m a t i c a l statistics) I n c l u d e s b i b l i o g r a p h i c a l r e f e r e n c e s a n d index. 1. S a m p l i n g (Statistics) I . T i t l e . QA276.6.C6 1977 001.4'222 77-728 ISBN G - 4 7 1 - I 6 2 4 D - X P r i n t e d in t h e U n i t e d S t a t e s o f A m e r i c a
30 29 28

Printed and b o o n d by Q u i n n . Woodbine. Inc.

12

MP

G TErn

0

il=
and

- 2.

(4.)=

da da

il =

1 0 !inti 3 2 d a a n d d a a r e call d 10 Itb a ing! timal! from a ingle u r v e , t h t a t e m o t 'IJ. Ii b e t w e e n da i n t certain t o b e correct. TIt % th a m s a m p l i n g p l a n r e u d many tim in a p o p u l a t i o n , a t a t m e n t b e i n g m a d fr m e a c h a m p l , a b o u t 9 9 % f th e t a t e m n auld r r e e t a n d 1% w r o n n sampling' in i o t r o d u c e d i n t an o p e r a t i n in hich mplete cem ha previ u Iy b e e n u s e d , a d e m o trati 0 f t h ' pr perty i m tim m a d e b d r a w i n g r e p e a t e d a m p l e f t h e t y p e p r o e d fr m a popuJati n f r which c o m p l e t r e c o r d s e ' t , t h a t IJ. i nown e e , e,g, ru bl nd e r t 1 5 7 . T h practical erificati n t h a t p p r o x i m a t e l lh t a t e d p r o p o r t i o n f t a t men i c o r r e c t d m u c h t o e d u c a t e a n d re u r dmin' trat a b o u t t h e n t u t e { ampling. imilarly w h e n a ingle a m p l e is ta e n from ch f a r i f differ o t p o p u l a t i o n u t 9 % f th 5% c o n f i d nce t a m e n a r e r r e c t . T h e p r c e d i n g d i cu ion urn t h a t (T~, m p u t e d from t h e a m p l e i kn wn exactly. etuall (T~,lik il i u b j e c t t o a m p l i o g e r r r. W i t h a normaJly di tri u t e d v a r i a b l e , t a b l e o f t u d e n t ' 1 d i s t r i b u t i o n a r e us d i n s t e a d o f t h e n rmal t a b l e to c a l c u l a t e confidenoe limi for IJ. w h e n t h e ampl i mall. R p l a c e m e n t o f t h e n o r m a l t a b l e by t h e 1 t a b l e m a k aIm t no difference i f t h e n u m b e r o f d e g r e e o f f r e e d o m in (Til e x c e e d s 50. W i t h c e r t a i n type o f tratified a m p l i n a n d with t h e m t h o d o f r e p l i c a t e d a m p l i n g ( e c t i o n 1 1 . 1 9 ) t h e d e g r e e o f f r e e d o m a r e mall a n d t h e 1 t a b l i n cd d .

Tc distri as sh kno\l freql
devi~

usin~

declll 1.96« WI

we cc wher must more

1.8 B
In
r

m p l e urvey t h e o r y i t i n.

d e timat

f r two Pu

f t h e m t m m o n p r o b l e m s , p a r t i c u l a r l in t h e t i m a t i o n t h a t a r e otherwise n eoient and uitabl a r e { u n d t n ' t h timat t h a t a r unbi c d in p r bability sampling, e r r f in th n u m b e t h a t we a r e a b l e me u r m nt n d n o n r e p o n e ma p r o d u c e b ' ( Q c o m p u t fr m t b e d a t a . T h i h a p p e n , f r in l a n c e , if t h e pc n h relu to b e i n t rvie e d a r e a1m t all o p ed t m e x p e n d i t u r e o f p u b l i c fun , w h e r e a tho e h o a r e j n t rvi e d a r e plit e n l f r n d ai t.

Thus

,

MPLI.'

TE

(I

in t h p r b a b i l i t t h a l t h i t h u n i t i d r s n i di t n u t d a b i n mi I n u m e r o f u c c e e H nc

1/
ut

a t e a h d r a . th a n a r e f n lri I 'Ih P = 1/ .

suits Yi t

(',) == n
J i n t i · Ih ariate

-.!-)( 1 -~)
o r lhi .
(.

a m u l t i n m i a l d i s l r i b u t i n.

T h e c(

in

2.3 · (2.

). n d 2 . 3 4 )
I[ ( y ) == n - I

ha

· f r amplin

ith r plac menl, 2.

==n

_,

.-1

(YI-

u_ I '1 -=-=--n n

_. 6)

quentl

mpling

ith ut repla

m nl i

Examr comprise populati( example, ratio o f a T h e sal the numt samples t R . I n larg negligiblt distributi,

Theore s a m p l e Oi approxim

where R '

2.11

a : A > . . . ., . . . . . .

TIO

Proof

I f n is larg o u t t h e dis x byX in

11

0 R NDOM

11'

ti n is o m t i m

4,
the

mi"(-U

=---n

(.7

The ,

und t rm

J j

;t

n i m de between. imple rand m 'th pr rti naJ a n d o p t i m u m II

nd

Thi

ed.
~)

means r e d n o t d b If t e r m s i n 1 /
~ ar

i n r e d r tati e 1 0 u n i t .

5.2 )
5)
h r th p t i m u m all a t i n i f r fixed n, t h a t i
2
I

w i l h nIt

Proof.
rart

=(l-f)n

(.9

in
,st

se
[fr m equation (. .
~d
I

n
ction .]

" /==-....:.;..n..:.:.... .
2

""

0)

'''''=--n--

6)

""

( .. I

Ifr m c q u a t i n

. 7 ) , . eti n

.

J