# plot integer data on a log-log scale

I'm new to statistics, and this may be a very trivial problem. I have an array of sorted data:

[31, 27, 26, 18, 11, 9, 8, 7, 6, 6, 6, 6, 5, 5, 5, 5, 5, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3]


and I want to plot it in a log-log scale. Then I decided to have in the x-axis the index in the array. Since the data is of integer values, I get the following curve:

Is there a better way to visualize this data?

• "Better" for what purpose? It's already fine for your stated objective, which is to "plot it in a log-log scale." – whuber Aug 19 '15 at 18:43
• Is this the way it's traditionally done when there are repeated values? I guess I'm asking what's the convention to plot this kind of data. – aaragon Aug 19 '15 at 18:44
• What information are you trying to convey with your plot? – whuber Aug 19 '15 at 18:46
• I'm trying to fit a line to that curve, and it seems to me that the repeating values are not helping. – aaragon Aug 19 '15 at 18:48
• What do you hope such a fitted line would tell you about the data? It would be related to the empirical cumulative distribution function, which points towards estimating parameters of the underlying population or distribution of which these data (presumably) are a sample. Is your objective to estimate that distribution, or is it to use it as a springboard for further analysis? Or do you really need a distribution estimate at all? – whuber Aug 19 '15 at 19:12