tection of common structures in multivariate data, e.g. in statistical process control. Critical issues are the choice of the number of principal components and their ...
!!"
# $ % "&'
( ) *)$+, - (#( ( $ . ( / 0 . Æ 1 )$+ *2# "34", 1 ( 5 . # . )$+( ( / #
) *)$+, - # # ( (#( 1 . . ( / # # - . 1 # . ( 2# *"34", ## )$+ - 0 # ( + . . *)$6 ,( 0 # # # 7 . Æ 1 - ( . )$6 ( 5 # 0 7 ) # 1 - ( 7 1 # ( / . . . # . . . 7# )$6 . * # ( "33 $ !88",( / . . 9 . .0# 01 # . ( 5 )$+ # *$7 5 "33: !888, # . )$+(
) # # . 1 ( 1 .
. . # . # # * , 7 ; / < 1 9 7 # < " #
. # # #
( 9 )$6 < * , . 1 . ( )$+ 9 . . 1 ( 2# *"34" 3, ## )$+ *", *, 0 . 1 # #
# *, ; ( 1 = ( 5 )$6 . *> >, )$+ .
)$6 ## . 2# 9 *, < * ,* ,
( ( . 1 * , 9 ( 5 # 7 . # 9 *, < * ,* ,
/ 9 =# - 7 # . " *, 7* , * , < " *, 7* , * , < ! ! *, (( *, < *, < *, ( *, 1 *, # # # *, 7 *, ?8 ! @ *, ( / -
? * ,
*,@ ? *,
* ,@ <
*,
/ # ( / )$+ 1 )$+ - ( 7 1 # # 11( )$+ (#( > > *7 , . 1 . ( + . 7 )$6 ( / )$6 A . % #6
*, *, )$6 #
- *, < **, *,, **, *,,( #9 # . # 1 * , 1 # ( . . # # ( )$6 #( %1 )$6 . . . ( # )$6 . . . # . 0 7 ) ( 5 10 . 1 # . 1 # .0# 01 # ( 1 1 . # )$6 1 1 . . 1 0# . . # . . 0 . # #9 1 . . 1# ( *!888, # . # )$+ )$+ # *+ "33&,( 9 0 . . . )$+
1 )$6 ( / "" # . . 9 )$6 ( %1 . 1 . . )$6 # # 7 . 1 )$6 ( # 1 )$6 . * # , . )$6 ( - 1 # ( / 0 . )$+ . . * (#( $. "34 ,( ) )$+ # . # 1 -
# . . ( %1 # ( 2 *"3:', ## . . ( # 7 # . ( % . )$+
)$6 . Æ . . 1 ( 0 ) 7 . Æ )$+ 1 - (
+ < * , 9
( . = . 1# 7 # 1 * , * , . # ( 5 7 1 * , * , ( # # .= . ( *!888,
1 # = . ( # . ( ( ( ) # 9 . 1 # # # . # ( ( * , * , ( # # . . * ,
1 7 # - # * , ( 1 . 1 # -
# . 7( ) # . 1 . ( / * , . 1 1 . * . . . ,
. ( ( < < " 7 # . 1 ( +
# 7 . 1 . ( + # # . 1 # # . 1 1 1 # . ( % # ( - 1 0 7 # - ( #. 0 . 1 . . = # . # ( . * # , . # . ( / # . * , 1 . . * ,( / # . . . 1 . ( 7 9 1 "" + *", ( / . 9 # . ( *!888,( # " 1 # . # :88 . # # * B !888,( 1# ## . 2# *"33:, 1 # # # * !88!,( / # # . 1 ( . "8 "" # ( % . . 1 & : # # 1 . 1 C 4 )$6 .( + . 1 # 7 .( B7 # 1 1 1 . C88 . . "" ( / > > # # C888 . 1 # "( # # 1
X1
X1
X5
X5
X2
X6
X2
X6 X3
X3
X7
X7 X4
X4
X8
X8 X10
X9
X10
X9
Partial correlation weak medium strong
X11
Partial correlation weak medium strong
X11
7 )$6 . # ( / # 1 7 # 01 # .
7 # ( # ! # . 1 9 . - ( # >> 0 + *", 1 D . # 1 ( % # )$ . 1 9 )$6
. # 1 9 #( / )$6 1 0 # . ( +
9 )$ # . . 1 7 )$6 ( 1 1 # . # 7# . 1 . ( %1 7 1 . Æ (
) # 1 # # # 1 ( . . . 1 > > . ( / . )$6 7 . ( 1 9 . . 0 .01 )$+ . )$6 ( 7 # 7 . # . # . >> )$6 ( E # . 1 . 1 # ( )$+ 1 . 1 . )$+( %1
-40
-40
-20
-20
0
0
20
20
40
40
60
! " #
$ % $ & "
'
0
100
200 Time
300
0
100
200
300
Time
)$+ . # . )$+( / )$+ Æ . */ ) F )0= !888,( E 1 + Æ . 1 . ( - . . (
2 B((( (( (5( *"3:',( / # . ( &C'&::( 2# ( ( *"34",( ; % ( 2# ( ( *"33:,( 0 # # ( "!&( $ )( *!88",( + # = . (
"' !4( $7 ( ( 5 G( *"33:,( ; $ H %( ( *!888,( / ( "C'"'!( ( B ( *!888,( )# . ;II111( .( J . #( IA I#( I ( ( ( *!88!,( #(
( ( *!888,( ! "# $ %& * ,( . /
( # K(( K 0 $( $( ( *"33 ,( ) # # . .0 ) ( ' ( 4!:4&4( $. ()( *"34 ,( ) . (
"&'" ( )(( *"33&,( ) # # ( ) &C&&:C( / (( ) F ( )0= +(B( *!888,( E ( '4348 (