Tukey did show that median and quantiles are often much more stable
than average.( f.e median polish and median smoothing in general)


-------------------
> We are solving a problem which should not be solved. The advantage
of Median and Quantiles, as opposed to Arithmetical Mean and Standard
Deviation, is that they can be evaluated without computing. Having a
computer at hand, this argument vanishes. 
> 
> --- Den søn 18/10/09 skrev Fraser Jackson
<[email protected]>:
> 
> > Fra: Fraser Jackson <[email protected]>
> > Emne: Re: [Jprogramming] "median" considered inaccurate?
> > Til: "Programming forum" <[email protected]>
> > Dato: søndag 18. oktober 2009 11.51
> > Statistical programs have a range of
> > alternatives for the quantile function. 
> > The following script embodies forms
> > considered in a useful survey paper some years ago. It
> > requires further 
> > functions to answer Devon's question but
> > does include some options worth considering.
> > 
> > NB.  Quantile functions
> > 
> > 
> > NB.  Each of the functions below will generate a plot
> > NB.  object which enables you to plot the quantile
> > function
> > NB.  for a data vector.  The plot object consists
> > of the
> > NB.  boxed p values and boxed quantiles.
> > 
> > NB.  Using this form linear interpolation
> > NB.  is used to find values between the data points in
> > the
> > NB.  plot object.
> > 
> > NB.  The functions are numbered as in Rob J. Hyndman
> > and Yanan Fan,
> > NB.  Sample Quantiles in Statistical Packages. 
> > The American
> > NB.  Statistician, 1996, Vol 50(4), 361-365. 
> > Their functions
> > NB.  QP2 and QP3 are closely related to QP1 and
> > dominated in nearly
> > NB.  all respects by later alternatives so are not
> > included.
> > 
> > NB.  I have checked the usage in SPlus 6 and R but
> > comments on
> > NB.  other packages are from Hyndman and Fan.
> > 
> > NB.  In Hyndman and Fan the treatment of p outside the
> > interval
> > NB.  associated with x(1),...,x(n) is not always
> > defined.  We have
> > NB.  adopted the uniform practise that in any such
> > region the
> > NB.  inverse of the EDF is used as the definition.
> > 
> > NB.  Classical definition - invert the EDF
> > 
> > NB.   Functions for Generating Graphs
> > NB.  [lowest limit, all upper limits] EDF frequency
> > 
> > EDF=: 3 : 0
> > (2#/:~y.);}.}:(2#i.>:#y.)%(#y.)
> > :
> > (}.2#y.,+`-/_1 _1 _2{y.); }:2#0,((+/\x.)%+/x.),1
> > )
> > 
> > NB.  Available in SAS  PROC UNIVARIATE
> > Invert =: |."1
> > QP1 =: [: Invert EDF
> > 
> > NB.  Parzen (1979)  Interpolates step function of
> > QP1
> > NB.  Available in SAS  PROC UNIVARIATE
> > QP4 =: ((([: i. [: >: #)%#);{.,])@/:~
> > 
> > NB.  Old definition proposed by Hazen(1914). 
> > Used by hydrologists.
> > NB.  Used in $tab interpolate in GLIM V3.77
> > NB.  Appears to be also used by SPlus and R for
> > estimating quartiles
> > NB.  for box plots.
> > QP5 =: 3 : 0
> > (0,(((i.n)+0.5)%n=. #y),1);({.,],{:)y =. /:~y.
> > )
> > 
> > NB.  Weibull(1939) and Gumbell(1939) both proposed
> > this measure.
> > NB.  Divides space into n+1 regions each with prob
> > %(n+1) on average.
> > NB.  Used in BMDP for quartiles.
> > NB.  Used in Minitab  DESCRIBE command for
> > quartiles.
> > NB.  Available in SAS  PROC UNIVARIATE
> > NB.  Appears to be used by SPSS
> > QP6 =: 3 : 0
> > (0,((>:i.n)%(1+n=.#y)),1);({.,],{:)y=. /:~y.
> > )
> > 
> > NB.  Gumbell(1939) also proposed this definition.
> > NB.  Divides the range in (n-1) intervals.
> > NB.  Exactly 100p% of intervals lie to the left of
> > QP7(p)
> > NB.  Used in SPlus 6  quantile(), but not for box
> > plots.
> > QP7 =: 3 : 0
> > ((i.#y)%<:#y);y =. /:~ y.
> > )
> > 
> > NB.  Reiss(1989) and Hyndman and Fan(1996)
> > NB.  Sample quantile is median unbiased of O(n^_0.5)
> > QP8 =: 3 : 0
> > (0,((_1r3+>:i.n)%(1r3+n=.#y)),1);({.,],{:)y =. /:~y.
> > )
> > 
> > NB.  Blom(1958) shows this is a better approximation
> > to
> > NB.  F(E(X(k))) for the normal distribution. 
> > QP9(p(k)) is
> > NB.  an approximately unbiased estimator of Q(p(k))
> > when F
> > NB.  is normal.  Tends to be used for normal QQ
> > plots.
> > QP9 =: 3 : 0
> > (0,((_0.375+>:i.n)%(0.25+n=.#y )),1);({.,],{:)y =.
> > /:~y.
> > )
> > 
> > QP =: Q8   NB.  The Hyndman and Fan
> > recommendation
> >            NB. 
> > Also used in R
> > 
> > 
> > ----- Original Message ----- 
> > From: "Sherlock, Ric" <[email protected]>
> > To: "Programming forum" <[email protected]>
> > Sent: Sunday, October 18, 2009 9:01 PM
> > Subject: Re: [Jprogramming] "median" considered
> > inaccurate?
> > 
> > 
> > > There is a problem with the previous general versions
> > and Don's original 
> > > for odd-numbered groups - ntiles2 fixes using the same
> > mechanism and 
> > > similar structure to median
> > >
> > >   ntiles1=: [: -:@(+/) (,: <:)@((%~
> > i.&.<:)@[ >.@:* #...@]) { /:~...@]
> > >
> > >   midpts=: (%~ i.&.<:)@[ *
> > <:@#...@]
> > >   ntiles2=: -:@(+/)@(<. ,:
> > >.)@midpts { /:~...@]
> > >
> > >   midpt=: -:@<:@#
> > >   median=: -:@(+/)@((<. ,
> > >.)@midpt { /:~)
> > >
> > >   median 3 4 5 6
> > > 4.5
> > >   2 ntiles1 3 4 5 6
> > > 4.5
> > >   2 ntiles2 3 4 5 6
> > > 4.5
> > >   median 3 4 5 6 7
> > > 5
> > >   2 ntiles1 3 4 5 6 7
> > > 5.5
> > >   2 ntiles2 3 4 5 6 7
> > > 5
> > >
> > >
> > >> -----Original Message-----
> > >> From: [email protected]
> > [mailto:programming-
> > >> [email protected]]
> > On Behalf Of Devon McCormick
> > >> Sent: Sunday, 18 October 2009 15:24
> > >> To: Programming forum
> > >> Subject: Re: [Jprogramming] "median" considered
> > inaccurate?
> > >>
> > >> Cool!  Consider it swiped!
> > >>
> > >> Quantiles are a very useful way to compare
> > stochastic models, e.g.
> > >> what's
> > >> the performance of bottom-decile PE stocks versus
> > top-decile ones?  And
> > >> if
> > >> there is a consistent relation between the top and
> > bottom deciles, does
> > >> it
> > >> also hold if we use 11-tiles or 9-tiles?
> > >>
> > >> On Sat, Oct 17, 2009 at 8:24 PM, Sherlock, Ric
> > >> <[email protected]>wrote:
> > >>
> > >> >   ntiles=: -:@(+/)@(] {~ (,:
> > <:)@([ ((%~ i.&.<:)@[ >.@:* #...@]) /:~...@]))
> > >> >
> > >> >   2 tiles scrs
> > >> > 61
> > >> >   3 tiles scrs
> > >> > 57 69
> > >> >   4 tiles scrs
> > >> > 52.5 61 70.5
> > >> >    5 tiles scrs
> > >> > 51 58.5 65.5 72.5
> > >> >
> > >> >
> > >> > > From: Devon McCormick
> > >> > >
> > >> > > If you can't stop, you should look to
> > generalize this: quartiles
> > >> are
> > >> > > only a
> > >> > > special case of N-tiles.
> > >> > >
> > >> > > On Sat, Oct 17, 2009 at 2:11 AM,
> > Sherlock, Ric
> > >> > > <[email protected]>wrote:
> > >> > >
> > >> > > > Sorry, couldn't stop....
> > >> > > >
> > >> > > > A few more versions of quartiles:
> > >> > > >
> > >> > > > Tidied up version of Don's
> > >> > > > quartiles0=: -:@(+/)@({~ (,:
> > <:)@(0.25 0.5 0.75 >.@:* #))@/:~
> > >> > > >
> > >> > > > Simplified version of Keith's
> > >> > > > quartiles1=: median (([: median ]
> > #~ >) , [ , [: median ] #~ <) ]
> > >> > > >
> > >> > > > A slightly different approach:
> > >> > > > quartiles2=: /:~@(median ([ , >
> > median/. ]) ])
> > >> > > >
> > >> > > >
> > >> > > > > From: Sherlock, Ric
> > >> > > > >
> > >> > > > > The following is based on
> > Keith Similie's stats companion.
> > >> > > > >
> > >> > > > > NB. Median and quartiles
> > >> > > > > midpt=: -:@<:@#
> > >> > > > > median=:
> > -:@(+/)@((<.,>.)@midpt { /:~)
> > >> > > > > Q1=: [: median ] #~ median
> > > ]
> > >> > > > > Q3=: [: median ] #~ median
> > < ]
> > >> > > > > quartiles=: Q1 , median , Q3
> > >> > > > >
> > >> > > > > Another definition of median
> > where the domain is integers.
> > >> > > > >
> > >> > > > > median=:
> > ~.@((<.,>.)@midpt { /:~)
> > >> > > > >
> > >> > > > > > From: Devon McCormick
> > >> > > > > >
> > >> > > > > > Don - I like yours better
> > than the one I have now, though
> > >> I'll
> > >> > > > > probably
> > >> > > > > > generalize it into an
> > "Ntiler".
> > >> > > > > >
> > >> > > > > > Part of the problem is
> > that there are multiple correct
> > >> answers if
> > >> > > we
> > >> > > > > > define
> > >> > > > > > quartile numbers as those
> > which divide the set as evenly as
> > >> > > possible
> > >> > > > > > into
> > >> > > > > > four groups, e.g.
> > >> > > > > >
> > >> > > > > >   
> > quartileCt=:  4 : '+/"1 (y>:/~x,_) *. y<
> > /~__,x'  NB.
> > >> Count
> > >> > > > > > elements/quartile
> > >> > > > > > NB. All these different
> > answers work correctly:
> > >> > > > > >    (52.75 61
> > 70.25) quartileCt scrs  NB. Excel
> > >> > > > > > 5 5 5 5
> > >> > > > > >    (52.5 61
> > 70.5) quartileCt scrs    NB. web site
> > >> > > > > > 5 5 5 5
> > >> > > > > >    (52.1 61.1
> > 70.1) quartileCt scrs  NB. another answer...
> > >> > > > > > 5 5 5 5
> > >> > > > > >
> > >> > > > > > One way to test, as you
> > suggest is to look at the behavior
> > >> when
> > >> > > we
> > >> > > > > have
> > >> > > > > > an
> > >> > > > > > odd number of elements,
> > i.e. "odd" with respect to four:
> > >> > > > > >
> > >> > > > > > NB. Two different ways of
> > counting number of
> > >> elements/quartile:
> > >> > > > > >   
> > quartileCt=:  4 : '+/"1 (y>:/~x,_) *. y< /~__,x'
> > >> > > > > >   
> > quartileCt2=: 4 : '+/"1 (y> /~x,_) *. y<:/~__,x'
> > >> > > > > > NB. Two different
> > quartilers:
> > >> > > > > >    test0=: 1 :
> > '(3{.4 ntilebps y) u y'  NB. Mine
> > >> > > > > >    test1=: 1 :
> > '(qr y) u y'         
> >    NB. Don's
> > >> > > > > >
> > >> > > > > > NB. Both work OK for even
> > and odd cases counted one way...
> > >> > > > > >    quartileCt
> > test0&>0 1 2 3 4}.&.><scrs
> > >> > > > > > 5 5 5 5
> > >> > > > > > 4 5 5 5
> > >> > > > > > 4 5 4 5
> > >> > > > > > 4 4 4 5
> > >> > > > > > 4 4 4 4
> > >> > > > > >    quartileCt
> > test1&>0 1 2 3 4}.&.><scrs
> > >> > > > > > 5 5 5 5
> > >> > > > > > 5 5 5 4
> > >> > > > > > 5 4 5 4
> > >> > > > > > 5 4 4 4
> > >> > > > > > 4 4 4 4
> > >> > > > > >
> > >> > > > > > NB. Mine falls down for a
> > couple of cases counted the other
> > >> way:
> > >> > > > > >    quartileCt2
> > test0&>0 1 2 3 4}.&.><scrs
> > >> > > > > > 4 5 5 6
> > >> > > > > > 4 5 5 5
> > >> > > > > > 4 4 5 5
> > >> > > > > > 4 4 4 5
> > >> > > > > > 3 4 4 5
> > >> > > > > > NB. but Don's works OK
> > under different counting method as
> > >> well:
> > >> > > > > >    quartileCt2
> > test1&>0 1 2 3 4}.&.><scrs
> > >> > > > > > 5 5 5 5
> > >> > > > > > 5 5 5 4
> > >> > > > > > 5 4 5 4
> > >> > > > > > 5 4 4 4
> > >> > > > > > 4 4 4 4
> > >> > > > > >
> > >> > > > > > Thanks for your
> > suggestions.
> > >> > > > > >
> > >> > > > > > Regards,
> > >> > > > > >
> > >> > > > > > Devon
> > >> > > > > >
> > >> > > > > > On Fri, Oct 16, 2009 at
> > 3:47 PM, Don Guinn
> > >> <[email protected]>
> > >> > > > > wrote:
> > >> > > > > >
> > >> > > > > > > Looked up the
> > definition of "median" and it appears that
> > >> there
> > >> > > are
> > >> > > > > > several
> > >> > > > > > > definitions of
> > "median". And, according to
> > >> > > > > > > http://en.wikipedia.org/wiki/Median median and
> > quartiles
> > >> can be
> > >> > > > > messy
> > >> > > > > > with
> > >> > > > > > > badly skewed data.
> > Best I can tell this is a measurement
> > >> that
> > >> > > > > should
> > >> > > > > > be
> > >> > > > > > > used
> > >> > > > > > > with care.
> > >> > > > > > > I wrote a quick verb
> > which gives the same answers as the
> > >> site
> > >> > > you
> > >> > > > > > > referenced
> > >> > > > > > > and it does strange
> > things, depending on the data. If the
> > >> count
> > >> > > of
> > >> > > > > > the set
> > >> > > > > > > is odd, which group
> > should have the extra number? What if
> > >> the
> > >> > > data
> > >> > > > > is
> > >> > > > > > > really
> > >> > > > > > > skewed?
> > >> > > > > > >
> > >> > > > > >
> > >   qr=.([:([:(+/%#)]{~[:(<:,:])[:>.0.25
> > 0.5 0.75"_*#)]/:])
> > >> NB.
> > >> > > Needs
> > >> > > > > > > cleaning up.
> > >> > > > > > >   qr
> > scrs
> > >> > > > > > > 52.5 61 70.5
> > >> > > > > > >    qr i.4
> > >> > > > > > > 0.5 1.5 2.5
> > >> > > > > > >   qr
> > i.5
> > >> > > > > > > 1.5 2.5 3.5
> > >> > > > > > >   qr
> > i.12
> > >> > > > > > > 2.5 5.5 8.5
> > >> > > > > > >   qr
> > i.11
> > >> > > > > > > 2.5 5.5 8.5
> > >> > > > > > >   qr
> > i.13
> > >> > > > > > > 3.5 6.5 9.5
> > >> > > > > >
> > >   -~/0 2{qr scrs
> > >> > > > > > > 18
> > >> > > > > > >   qr
> > 1 1 1 1 1 2 3 4
> > >> > > > > > > 1 1 2.5
> > >> > > > > > >
> > >> > > > > > >
> > >> > > > > > > On Fri, Oct 16, 2009
> > at 1:21 PM, Devon McCormick
> > >> > > > > <[email protected]>
> > >> > > > > > > wrote:
> > >> > > > > > >
> > >> > > > > > > > Members of the
> > forum -
> > >> > > > > > > >
> > >> > > > > > > > while looking
> > up some statistical definitions, I came
> > >> across
> > >> > > this
> > >> > > > > > example
> > >> > > > > > > >
> > >> > > > > >
> > >> > > > >
> > >> > >
> > >>
http://www2.le.ac.uk/offices/ssds/sd/ld/resources/numeracy/variability
> > >> > > > > > > > in which the
> > calculation of the median disagrees with the
> > >> > > result
> > >> > > > > of
> > >> > > > > > the
> > >> > > > > > > one
> > >> > > > > > > > listed as "m0=:
> > median=: <....@-:@# { /:~" in "MathStats" on
> > >> the
> > >> > > J
> > >> > > > > > wiki.
> > >> > > > > > > >
> > >> > > > > > > > I was actually
> > looking at the definition of quartiles
> > >> when I
> > >> > > > > > noticed
> > >> > > > > > > this.
> > >> > > > > > > >
> > >> > > > > > > > For the series
> > >> > > > > > > >
> > >> > > > > > >
> > >   #scrs=. 43 48 50 50 52 53 56 58 59 60
> > 62 65 66 68 70 71
> > >> 74
> > >> > > 76
> > >> > > > > 78
> > >> > > > > > 80
> > >> > > > > > > > 20
> > >> > > > > > >
> > >   m0=: <....@-:@# { /:~
> > >> > > > > > >
> > >   m0 scrs
> > >> > > > > > > > 62
> > >> > > > > > >
> > >   median scrs  NB. my own
> > definition
> > >> > > > > > > > 61
> > >> > > > > > >
> > >   median
> > >> > > > > > > > -:@(+/)@((<.
> > , >.)@midpt { /:~)
> > >> > > > > > >
> > >   midpt
> > >> > > > > > > > -:@<:@#
> > >> > > > > > > >
> > >> > > > > > > > Also, this
> > site's answers disagree with Excel and with my
> > >> own
> > >> > > > > > quartile
> > >> > > > > > > > function,
> > applied to "scrs" above, but I think the site
> > >> is
> > >> > > > > correct:
> > >> > > > > > >
> > >   NB. Quartiles 1-3 according to Excel:
> > >> > > > > > >
> > >   52.75 61 70.25
> > >> > > > > > > >
> > >> > > > > > >
> > >   NB. According to
> > >> > > > > > > >
> > >> > > > > >
> > >> > > > >
> > >> > >
> > >>
http://www2.le.ac.uk/offices/ssds/sd/ld/resources/numeracy/variability
:
> > >> > > > > > >
> > >   52.5 61 70.5
> > >> > > > > > > >
> > >> > > > > > >
> > >   0 1 2 quartile&><scrs
> > >> > > > > > > > 52 60 70
> > >> > > > > > > >
> > >> > > > > > > > NB. My
> > "quartile" disagrees with my "median": the middle
> > >> > > quartile
> > >> > > > > > should
> > >> > > > > > > be
> > >> > > > > > > > the same as the
> > median.
> > >> > > > > > >
> > >   quartile
> > >> > > > > > > > 4 : 'x{4
> > ntilebps y'
> > >> > > > > > >
> > >   ntilebps
> > >> > > > > > > > 4 : 0
> > >> > > > > > > > NB.* ntilebps:
> > return breakpoint values of x-tiles of y;
> > >> e.g.
> > >> > > 4
> > >> > > > > > ntilebps
> > >> > > > > > > y
> > >> > > > > > > > NB.  ->
> > quartiles; 0-based so "1st" quartile is 0{4
> > >> ntilebps
> > >> > > y.
> > >> > > > > > >
> > >   quant=. x
> > >> > > > > > >
> > >   y=. /:~y
> > >> > > > > > >
> > >   wh=. 0 1#:(i.quant)*quant%~#y 
> > NB. Where partition
> > >> points
> > >> > > are
> > >> > > > > > exactly
> > >> > > > > > >
> > >   'n f'=. |:wh     
> >               NB. whole
> > and
> > >> fractional
> > >> > > part
> > >> > > > > of
> > >> > > > > > > > partitions
> > >> > > > > > >
> > >   1|.+/"1 ((1-f),.f)*(n+/_1 0){y NB.
> > "1|." moves top
> > >> quantile
> > >> > > to
> > >> > > > > > end.
> > >> > > > > > > > )
> > >> > > > > > > >
> > >> > > > > > > > Anyone care to
> > weigh in on this?
> > >> > > > > > > >
> > >> > > > > > > > Regards,
> > >> > > > > > > >
> > >> > > > > > > > Devon
> > >> > > > > > > >
> > >> > > > > > > >
> > >> > > > > > > > --
> > >> > > > > > > > Devon
> > McCormick, CFA
> > >> > > > > > > > ^me^ at acm.
> > >> > > > > > > > org is my
> > >> > > > > > > > preferred
> > e-mail
> > >> > > > > > > >
> > ---------------------------------------------------------
> > >> ----
> > >> > > ----
> > >> > > > > --
> > >> > > > > > ---
> > >> > > > > > > > For information
> > about J forums see
> > >> > > > > > http://www.jsoftware.com/forums.htm
> > >> > > > > > > >
> > >> > > > > > >
> > -----------------------------------------------------------
> > >> ----
> > >> > > ----
> > >> > > > > --
> > >> > > > > > -
> > >> > > > > > > For information
> > about J forums see
> > >> > > > > > http://www.jsoftware.com/forums.htm
> > >> > > > > > >
> > >> > > > > >
> > >> > > > > >
> > >> > > > > >
> > >> > > > > > --
> > >> > > > > > Devon McCormick, CFA
> > >> > > > > > ^me^ at acm.
> > >> > > > > > org is my
> > >> > > > > > preferred e-mail
> > >> > > > > >
> > -------------------------------------------------------------
> > >> ----
> > >> > > ----
> > >> > > > > -
> > >> > > > > > For information about J
> > forums see
> > >> > > > > http://www.jsoftware.com/forums.htm
> > >> > > > >
> > ---------------------------------------------------------------
> > >> ----
> > >> > > ---
> > >> > > > > For information about J forums
> > see
> > >> > > http://www.jsoftware.com/forums.htm
> > >> > > >
> > -----------------------------------------------------------------
> > >> ----
> > >> > > -
> > >> > > > For information about J forums see
> > >> > > http://www.jsoftware.com/forums.htm
> > >> > > >
> > >> > >
> > >> > >
> > >> > >
> > >> > > --
> > >> > > Devon McCormick, CFA
> > >> > > ^me^ at acm.
> > >> > > org is my
> > >> > > preferred e-mail
> > >> > >
> >
-------------------------------------------------------------------
> > >> ---
> > >> > > For information about J forums see
> > >> http://www.jsoftware.com/forums.htm
> > >> >
> >
---------------------------------------------------------------------
> > >> -
> > >> > For information about J forums see
> > >> http://www.jsoftware.com/forums.htm
> > >> >
> > >>
> > >>
> > >>
> > >> --
> > >> Devon McCormick, CFA
> > >> ^me^ at acm.
> > >> org is my
> > >> preferred e-mail
> > >>
> >
----------------------------------------------------------------------
> > >> For information about J forums see
http://www.jsoftware.com/forums.htm
> > >
> >
----------------------------------------------------------------------
> > > For information about J forums see
http://www.jsoftware.com/forums.htm 
> > 
> > 
> >
----------------------------------------------------------------------
> > For information about J forums see
http://www.jsoftware.com/forums.htm
> > 
> 
> 
>       Trænger du til at se det store billede? Kelkoo giver dig gode
tilbud på LCD TV! Se her http://dk.yahoo.com/r/pat/lcd
>
----------------------------------------------------------------------
> For information about J forums see
http://www.jsoftware.com/forums.htm
> 
> 
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to