Date: Fri, 26 Jun 2009 17:11:53 +0200
Reply-To: Marta García-Granero <mgarciagranero@gmail.com>
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: Marta García-Granero <mgarciagranero@gmail.com>
Subject: Re: Jarque-Bera test
In-Reply-To: <cad0d1200906240632u2a68e264j63bf0884aafb3d31@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Hi again Antonio:
When I replied to your request concerning Jarque-bera test (see my
message at the end of this one), I was a bit busy, and didn't extend on
the answer.
Jarque-Bera test is a normality test that combines skewness and kurtosis
into one chi-square statistic with 2 df. Therefore, it is not a
"symmetry around the median" test, as you were looking for. My syntax IS
really a test for symmetry around the median, a modification of the one
sample Wilcoxon test, as described by Jerrold Zar (in his book:
Biostatistical Analysis).
HTH,
Marta GG
--------------------------
My message:
>
> Anybody know how can I do a Jarque-Bera skewness test with spss?
>
> what I want to see is if there're symmetry around the median,
>
>
> Hi:
>
> See if this syntax helps (it computes non parametric masures of
> skewness
> and kurtosis and performs a one sample symmetry test (based on
> Wilcoxon
> test)
>
> * Sample dataset *.
> DATA LIST FREE/bilirrubin (F8).
> BEGIN DATA
> 25 178 15 10 90 30 280 20 58 8
> 20 75 85 12 158 10 6 22 103 16
> 170 28 240 8 16 20 80 420 36 24
> 70 22 14 5 130 54 25 22 15 22
> 50 24 143 42 4 18 44 220 54 38
> 135 24 78 24 4 152 68 45 38 18
> 120 18 30 20 360 26 12 16 310 72
> 48 96 32 22 55 12 10 62 46 35
> 15 192 20 65 42 6 60 34 14 115
> 9 164 12 40 18 530 32 104 28 6
> 81 10 26 94 46 40 14 40
> END DATA.
> VAR LEVEL bilirrubin(SCALE).
>
> * Syntax *.
> OMS
> /SELECT TABLES
> /IF COMMANDS = 'Explore'
> SUBTYPES = 'Percentiles'
> /DESTINATION FORMAT = SAV
> OUTFILE = 'C:\Temp\Octiles.sav'.
> OMS
> /SELECT TABLES
> /IF COMMANDS = 'Explore'
> SUBTYPES = 'Case Processing Summary'
> /DESTINATION VIEWER = NO.
> EXAMINE
> VARIABLES=bilirrubin
> /PLOT BOXPLOT
> /PERCENTILES(12.5,25,37.5,50,62.5,75,87.5) AEMPIRICAL
> /STATISTICS NONE
> /NOTOTAL.
> OMSEND.
>
> MATRIX.
> * Get octiles *.
> GET Ei /VAR=@12.5 TO @87.5 /FILE='C:\Temp\Octiles.sav'
> /MISSING=OMIT.
> * Skewness: Bowley's SK2 *.
> COMPUTE SK2=(Ei(6)+Ei(2)-2*Ei(4))/(Ei(6)-Ei(2)).
> * Kurtosis: Moor's KR2 *.
> COMPUTE KR2=(Ei(7)-Ei(5)+Ei(3)-Ei(1))/(Ei(6)-Ei(2))-1.233.
> * Reports *.
> PRINT Ei(4)
> /FORMAT='F8.1'
> /TITLE='Sample median (used for Wilcoxon signed-rank test of symmetry
> around the median)'.
> PRINT {SK2;KR2}
> /FORMAT='F8.3'
> /RLABEL='SK2','KR2'
> /TITLE="Robust measures of skweness & kurtosis: Bowley's SK2 and
> Moor's
> centered KR2".
> PRINT/TITLE='SK2 Skewness: range [-1; 1]. -ve skewed to the left, +ve
> skewed to the right'.
> PRINT/TITLE='KR2 Kurtosis: range [-1.233; infinity]. -ve platikurtic,
> +ve leptokurtic'.
> END MATRIX.
>
> TEMPORARY.
> COMPUTE k=1.
> AGGREGATE
> /OUTFILE=*
> MODE=ADDVARIABLES
> /BREAK=k
> /SampleMedian = MEDIAN(bilirrubin).
> NPAR TEST
> /WILCOXON=SampleMedian WITH bilirrubin .
> DELETE VARIABLES SampleMedian.
>
> HTH,
> Marta
>
> --
> For miscellaneous SPSS related statistical stuff, visit:
> http://gjyp.nl/marta/
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> LISTSERV@LISTSERV.UGA.EDU <mailto:LISTSERV@LISTSERV.UGA.EDU> (not
> to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD
>
>
>
>
> --
> Alexander J. Shackman, Ph.D.
> Laboratory for Affective Neuroscience
> Waisman Laboratory for Brain Imaging & Behavior
> University of Wisconsin-Madison
> 1202 West Johnson Street
> Madison, Wisconsin 53706
>
> Telephone: +1 (608) 358-5025
> Fax: +1 (608) 265-2875
> Email: shackman@wisc.edu <mailto:shackman@wisc.edu>
> http://psyphz.psych.wisc.edu/~shackman
> <http://psyphz.psych.wisc.edu/%7Eshackman>
> Calendar {still under construction}:
> http://www.google.com/calendar/embed?src=ajshackman%40gmail.com
--
For miscellaneous SPSS related statistical stuff, visit:
http://gjyp.nl/marta/
=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@LISTSERV.UGA.EDU (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
| 
