let I, Omega be non empty set ; :: thesis: for Sigma being SigmaField of Omega
for P being Probability of Sigma
for F being ManySortedSigmaField of I,Sigma
for J, K being non empty Subset of I st F is_independent_wrt P & J misses K holds
for a, c being Subset of Omega st a in MeetSections J,F & c in MeetSections K,F holds
P . (a /\ c) = (P . a) * (P . c)
let Sigma be SigmaField of Omega; :: thesis: for P being Probability of Sigma
for F being ManySortedSigmaField of I,Sigma
for J, K being non empty Subset of I st F is_independent_wrt P & J misses K holds
for a, c being Subset of Omega st a in MeetSections J,F & c in MeetSections K,F holds
P . (a /\ c) = (P . a) * (P . c)
let P be Probability of Sigma; :: thesis: for F being ManySortedSigmaField of I,Sigma
for J, K being non empty Subset of I st F is_independent_wrt P & J misses K holds
for a, c being Subset of Omega st a in MeetSections J,F & c in MeetSections K,F holds
P . (a /\ c) = (P . a) * (P . c)
let F be ManySortedSigmaField of I,Sigma; :: thesis: for J, K being non empty Subset of I st F is_independent_wrt P & J misses K holds
for a, c being Subset of Omega st a in MeetSections J,F & c in MeetSections K,F holds
P . (a /\ c) = (P . a) * (P . c)
let J, K be non empty Subset of I; :: thesis: ( F is_independent_wrt P & J misses K implies for a, c being Subset of Omega st a in MeetSections J,F & c in MeetSections K,F holds
P . (a /\ c) = (P . a) * (P . c) )
assume A0:
F is_independent_wrt P
; :: thesis: ( not J misses K or for a, c being Subset of Omega st a in MeetSections J,F & c in MeetSections K,F holds
P . (a /\ c) = (P . a) * (P . c) )
assume A1:
J misses K
; :: thesis: for a, c being Subset of Omega st a in MeetSections J,F & c in MeetSections K,F holds
P . (a /\ c) = (P . a) * (P . c)
let a, c be Subset of Omega; :: thesis: ( a in MeetSections J,F & c in MeetSections K,F implies P . (a /\ c) = (P . a) * (P . c) )
assume A2:
( a in MeetSections J,F & c in MeetSections K,F )
; :: thesis: P . (a /\ c) = (P . a) * (P . c)
consider E1 being non empty finite Subset of I, f1 being SigmaSection of E1,F such that
A3:
( E1 c= J & a = meet (rng f1) )
by A2, Def9;
consider E2 being non empty finite Subset of I, f2 being SigmaSection of E2,F such that
A4:
( E2 c= K & c = meet (rng f2) )
by A2, Def9;
consider n being Nat such that
A5:
E1, Seg n are_equipotent
by FINSEQ_1:77;
consider m being Nat such that
A6:
E2, Seg m are_equipotent
by FINSEQ_1:77;
consider e1 being Function such that
A7:
( e1 is one-to-one & dom e1 = Seg n & rng e1 = E1 )
by A5, WELLORD2:def 4;
A8:
e1 <> {}
by A7;
consider e2 being Function such that
A9:
( e2 is one-to-one & dom e2 = Seg m & rng e2 = E2 )
by A6, WELLORD2:def 4;
A10:
e2 <> {}
by A9;
A11:
( f1 is_independent_wrt P & f2 is_independent_wrt P )
by A0, Def5;
reconsider rngf1 = rng f1, rngf2 = rng f2 as set ;
reconsider f1 = f1 as Function of E1,rngf1 by FUNCT_2:8;
reconsider f2 = f2 as Function of E2,rngf2 by FUNCT_2:8;
reconsider e1 = e1 as Function of Seg n,E1 by A7, FUNCT_2:3;
reconsider e2 = e2 as Function of Seg m,E2 by A9, FUNCT_2:3;
A12:
( rng (f1 * e1) = rng f1 & rng (f2 * e2) = rng f2 )
by A7, A9, FUNCT_2:20;
A13:
( e1 is FinSequence & e2 is FinSequence )
by A7, A9, FINSEQ_1:def 2;
then reconsider e1 = e1 as one-to-one FinSequence of E1 by A7, FINSEQ_1:def 4;
reconsider e2 = e2 as one-to-one FinSequence of E2 by A9, A13, FINSEQ_1:def 4;
reconsider Si = Sigma as set ;
reconsider f1 = f1 as Function of E1,Si ;
reconsider fe1 = f1 * e1 as FinSequence of Si ;
reconsider f2 = f2 as Function of E2,Si ;
reconsider fe2 = f2 * e2 as FinSequence of Si ;
reconsider Pfe1 = (P * f1) * e1, Pfe2 = (P * f2) * e2 as FinSequence of REAL by FINSEQ_2:36;
deffunc H1( set ) -> set = f1 . $1;
deffunc H2( set ) -> set = f2 . $1;
defpred S1[ set ] means $1 in E1;
consider h being Function such that
A14:
( dom h = E1 \/ E2 & ( for i being set st i in E1 \/ E2 holds
( ( S1[i] implies h . i = H1(i) ) & ( not S1[i] implies h . i = H2(i) ) ) ) )
from PARTFUN1:sch 1();
A15:
fe1 ^ fe2 = h * (e1 ^ e2)
proof
(
rng e1 = dom f1 &
rng e2 = dom f2 )
by A7, A9, FUNCT_2:def 1;
then B0:
(
len fe1 = len e1 &
len fe2 = len e2 )
by FINSEQ_2:33;
for
x being
set st
x in dom (h * (e1 ^ e2)) holds
x in dom (e1 ^ e2)
by FUNCT_1:21;
then B1:
dom (h * (e1 ^ e2)) c= dom (e1 ^ e2)
by TARSKI:def 3;
for
x being
set st
x in dom (e1 ^ e2) holds
x in dom (h * (e1 ^ e2))
then B2:
dom (e1 ^ e2) c= dom (h * (e1 ^ e2))
by TARSKI:def 3;
B3:
dom (fe1 ^ fe2) =
Seg ((len fe1) + (len fe2))
by FINSEQ_1:def 7
.=
dom (e1 ^ e2)
by FINSEQ_1:def 7, B0
.=
dom (h * (e1 ^ e2))
by B1, B2, XBOOLE_0:def 10
;
for
x being
set st
x in dom (fe1 ^ fe2) holds
(fe1 ^ fe2) . x = (h * (e1 ^ e2)) . x
proof
let x be
set ;
:: thesis: ( x in dom (fe1 ^ fe2) implies (fe1 ^ fe2) . x = (h * (e1 ^ e2)) . x )
assume C0:
x in dom (fe1 ^ fe2)
;
:: thesis: (fe1 ^ fe2) . x = (h * (e1 ^ e2)) . x
then reconsider x =
x as
Element of
NAT ;
per cases
( x in dom fe1 or not x in dom fe1 )
;
suppose
not
x in dom fe1
;
:: thesis: (fe1 ^ fe2) . x = (h * (e1 ^ e2)) . xthen consider n being
Nat such that F0:
(
n in dom fe2 &
x = (len fe1) + n )
by C0, FINSEQ_1:38;
F1:
n in dom e2
by F0, FUNCT_1:21;
then F2:
e2 . n in E2
by A9, FUNCT_1:12;
E1 /\ E2 c= J /\ K
by A3, A4, XBOOLE_1:27;
then
E1 /\ E2 c= {}
by A1, XBOOLE_0:def 7;
then
E1 /\ E2 = {}
;
then
E2 = (E2 \ E1) \/ {}
by XBOOLE_1:51;
then F3:
not
e2 . n in E1
by F2, XBOOLE_0:def 5;
F4:
E2 c= E1 \/ E2
by XBOOLE_1:7;
(fe1 ^ fe2) . x =
fe2 . n
by F0, FINSEQ_1:def 7
.=
f2 . (e2 . n)
by F0, FUNCT_1:22
.=
h . (e2 . n)
by F2, F3, F4, A14
.=
h . ((e1 ^ e2) . x)
by F0, F1, B0, FINSEQ_1:def 7
;
hence
(fe1 ^ fe2) . x = (h * (e1 ^ e2)) . x
by C0, B3, FUNCT_1:22;
:: thesis: verum end; end;
end;
hence
fe1 ^ fe2 = h * (e1 ^ e2)
by B3, FUNCT_1:9;
:: thesis: verum
end;
for i being set st i in E1 \/ E2 holds
h . i in Sigma
then reconsider h = h as Function of E1 \/ E2,Sigma by A14, FUNCT_2:5;
for i being set st i in E1 \/ E2 holds
h . i in F . i
then reconsider h = h as SigmaSection of E1 \/ E2,F by Def4;
A16:
h is_independent_wrt P
by Def5, A0;
E1 c= E1 \/ E2
by XBOOLE_1:7;
then
rng e1 c= E1 \/ E2
by XBOOLE_1:1;
then reconsider e1 = e1 as FinSequence of E1 \/ E2 by FINSEQ_1:def 4;
E2 c= E1 \/ E2
by XBOOLE_1:7;
then
rng e2 c= E1 \/ E2
by XBOOLE_1:1;
then reconsider e2 = e2 as FinSequence of E1 \/ E2 by FINSEQ_1:def 4;
E1 /\ E2 c= J /\ K
by A3, A4, XBOOLE_1:27;
then
E1 /\ E2 c= {}
by A1, XBOOLE_0:def 7;
then
E1 /\ E2 = {}
;
then
rng e1 misses rng e2
by A7, A9, XBOOLE_0:def 7;
then reconsider e12 = e1 ^ e2 as one-to-one FinSequence of E1 \/ E2 by FINSEQ_3:98;
reconsider e1 = e1 as one-to-one FinSequence of E1 ;
reconsider fe1 = fe1 as FinSequence of Si ;
reconsider e2 = e2 as FinSequence of E2 ;
reconsider fe2 = fe2 as FinSequence of Si ;
reconsider P = P as Function of Si, REAL ;
A18: (P * h) * e12 =
P * (h * (e1 ^ e2))
by RELAT_1:55
.=
(P * fe1) ^ (P * fe2)
by FINSEQOP:10, A15
;
A19:
( P * fe1 = Pfe1 & P * fe2 = Pfe2 )
by RELAT_1:55;
reconsider P = P as Function of Sigma, REAL ;
reconsider f1 = f1 as Function of E1,Sigma ;
A21:
Product ((P * f1) * e1) = P . (meet (rng (f1 * e1)))
by A8, A11, Def3;
reconsider e2 = e2 as one-to-one FinSequence of E2 ;
reconsider f2 = f2 as Function of E2,Sigma ;
P . (a /\ c) =
P . (meet ((rng f1) \/ (rng f2)))
by SETFAM_1:10, A3, A4
.=
P . (meet (rng (fe1 ^ fe2)))
by FINSEQ_1:44, A12
.=
Product (Pfe1 ^ Pfe2)
by A18, A19, A16, A8, Def3, A15
.=
(Product Pfe1) * (Product Pfe2)
by RVSUM_1:127
.=
(P . a) * (P . c)
by A3, A4, A12, A21, A10, A11, Def3
;
hence
P . (a /\ c) = (P . a) * (P . c)
; :: thesis: verum