[ncl-talk] stat_dispersion and sample size limit

[Rainer Luptowitz]

Hi ncl-talk

Is the "stat_dispersion" ncl function suppose to output fill in values for
stats 3 to 28 if the population or sample size is less then 10?

Here is an output from "stat_dispersion" function where the sample "b" size
is 8

Variable: b
Type: float
Total Size: 32 bytes
            8 values
Number of Dimensions: 1
Dimensions and sizes: [ncl_join | 8]  ; I had joined 8 files trends with
"addfiles" function
Coordinates:
Number Of Attributes: 0
(0) -8.627946
(1) -1.777657
(2) -0.09471972
(3) 0.05206749
(4) -1.341161
(5) -2.990715
(6) -0.3975176
(7) 0.4604226
(0)
(0)    ===> Robust Dispersion Statistics <===
(0)  [0]            Mean=-1.83965
(0)  [1]          StdDev=8.80739
(0)  [2]             Min=9.96921e+36
(0)  [3]          LowDec=9.96921e+36
(0)  [4]          LowOct=9.96921e+36
(0)  [5]          LowSex=9.96921e+36
(0)  [6]     LowQuartile=9.96921e+36
(0)  [7]          LowTri=9.96921e+36
(0)  [8]          Median=9.96921e+36
(0)  [9]         HighTri=9.96921e+36
(0)  [10]   HighQuartile=9.96921e+36
(0)  [11]        HighSex=9.96921e+36
(0)  [12]        HighOct=9.96921e+36
(0)  [13]        HighDec=9.96921e+36
(0)  [14]            Max=9.96921e+36
(0)  [15]          Range=9.96921e+36
(0)  [16]     Dispersion=9.96921e+36
(0)  [17]    RMS Anomaly=9.96921e+36
(0)  [18]      #   Total=8
(0)  [19]      #    Used=8
(0)  [20]      # Missing=0
(0)  [21]      % Missing=0
(0)  [22]     Lower 0.1%=9.96921e+36
(0)  [23]     Lower 1.0%=9.96921e+36
(0)  [24]     Lower 5.0%=9.96921e+36
(0)  [25]     Upper 5.0%=9.96921e+36
(0)  [26]     Upper 1.0%=9.96921e+36
(0)  [27]     Upper 0.1%=9.96921e+36
(0)  [28]       Skewness=-1.37234
(0)  [29]       Kurtosis=0.512403



What do you think?

Thank you for your time and assistance,

*Rainer Luptowitz*
*Grad Student *
*Dept Earth Sciences*
*UC Riverside*
*rlupt001 at ucr.edu* <rlupt001 at ucr.edu>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20150424/44c1def9/attachment.html