Re: histogram oddities

[John Darrington]

On Sun, Aug 06, 2017 at 11:57:37AM +0200, ftr public wrote:
     
     The GRAPH procedure gives the annexed histogram dp-1.png.
     The FREQUENCIES procedure gives the annexed histogram dp-1.png.
     
     What is surprising is that the two procedures give different graphs, and
     that in both the data are not truthfully mirrored by the visualisation.


     There should be no empty space between two values.
     
Why do you think that?  The bin width of this histogram is 0.4
So the first bin includes values in the range [0, 0.4) and from your frequency
table we can see there are 115 of them.  The second bin takes values in the 
range
[0.4, 0.8) and there are none.  So the histogram is drawn with a bar of zero 
height.
The next bin is [0.8, 1.2) and there are 36 such samples (all of which have the 
value
1).

This is why the histogram appears "gappy".  It's because you are presenting it 
with 
descrete data.  Histograms are normally used with continuous data.

J'




-- 
Avoid eavesdropping.  Send strong encrypted email.
PGP Public key ID: 1024D/2DE827B3 
fingerprint = 8797 A26D 0854 2EAB 0285  A290 8A67 719C 2DE8 27B3
See http://sks-keyservers.net or any PGP keyserver for public key.

signature.asc
Description: Digital signature