Comments on The distribution of online data usageTypePad2010-06-13T23:46:50ZBlog Administratorhttps://blog.revolutionanalytics.com/tag:typepad.com,2003:https://blog.revolutionanalytics.com/2010/06/the-distribution-of-online-data-usage/comments/atom.xml/Ralph Winters commented on 'The distribution of online data usage'tag:typepad.com,2003:6a010534b1db25970b0134849e96a4970c2010-06-18T19:55:27Z2010-06-18T19:55:27ZRalph WintersAgree with the above comments: This is not a normal distribution. The leftmost bound of x axis should be 0...<p>Agree with the above comments: This is not a normal distribution. The leftmost bound of x axis should be 0 (not negative) and using standard deviation of normal distribution is not correct. Need to use another distribution...</p>
<p>-Ralph Winters</p>Stephen commented on 'The distribution of online data usage'tag:typepad.com,2003:6a010534b1db25970b0133f15d863f970b2010-06-17T09:03:53Z2010-06-17T09:03:53ZStephenhttp://www.mendeley.com/profiles/stephen-henderson/hmmmm.... Its highly likely (ie its a regular mistake) that the top distribution is a fit to data with some...<p>hmmmm....</p>
<p>Its highly likely (ie its a regular mistake) that the top distribution is a fit to data with some super user peaks that are not apparent (should have loessed it). </p>
<p>So the 97 percentile is (probably) in the right place but the curve itself a poor representation.</p>N. N. commented on 'The distribution of online data usage'tag:typepad.com,2003:6a010534b1db25970b0133f1553ade970b2010-06-16T21:47:30Z2010-06-16T21:47:30ZN. N.Use "Empirical Cumulative Distribution Function", ecdf() or something instead. I use about 18G of mobile data monthly with 7 Mbit/s...<p>Use "Empirical Cumulative Distribution Function", ecdf() or something instead.</p>
<p>I use about 18G of mobile data monthly with 7 Mbit/s modem. New modems are 21 Mbit/s. Theoretically my plan is capped to 3G/month but the operator doesn't seem to care. This costs 13 euros in a month. <br />
</p>Rick commented on 'The distribution of online data usage'tag:typepad.com,2003:6a010534b1db25970b01348472052b970c2010-06-16T20:02:29Z2010-06-16T20:02:29ZRickYou can't model these data by a normal distribution. Note that your model implies that a large number of cellphone...<p>You can't model these data by a normal distribution. Note that your model implies that a large number of cellphone users use NEGATIVE megabytes. The real usage is bounded below by zero, but is unbounded above. Simple models for this situation include lognormal and gamma.</p>Nick N commented on 'The distribution of online data usage'tag:typepad.com,2003:6a010534b1db25970b013484708ffe970c2010-06-16T16:20:16Z2010-06-16T16:20:16ZNick NNot to mention that the probability of using less than 0 MB of data is 0. With an "average" of...<p>Not to mention that the probability of using less than 0 MB of data is 0. With an "average" of 200 MB and a standard deviation of 159 MB, a normal distribution is going to be significantly wrong. My guess is a gamma distribution, but others could make sense.</p>
<p>Also, I'd take the "average" of 200 MB as the mode, not the mean, given the graph.</p>John Christie commented on 'The distribution of online data usage'tag:typepad.com,2003:6a010534b1db25970b0133f1494b0a970b2010-06-16T16:09:11Z2010-06-16T16:09:11ZJohn ChristieIt would be great to see the data because this must not be normal and must be fat tailed (as...<p>It would be great to see the data because this must not be normal and must be fat tailed (as Bill stated) because with that SD then the 99% is less than 700, not the 800 in the chart.</p>Bill Mill commented on 'The distribution of online data usage'tag:typepad.com,2003:6a010534b1db25970b0133f1492e3a970b2010-06-16T15:46:26Z2010-06-16T15:46:26ZBill Millhttp://billmill.orgCellphone data usage is almost certainly a fat-tailed distribution, not a normal one. The second graph says absolutely nothing about...<p>Cellphone data usage is almost certainly a fat-tailed distribution, not a normal one. The second graph says absolutely nothing about the first, and shame on this article for promoting the idea that every distribution is normal.</p>
<p>I hate the cellphone companies too, but this is silly,<br />
Bill Mill</p>