Archive for the ‘Research’ Category
Review of Previous Parts
In Part 1, we described how two datasets can have strong, weak or no correlation depending on how the points in the resultant scatter plot are located relative to one another along a straight line. In keeping with conventional practice, the two variables are often referred to separately as X, with sample values x1, x2, … , xn, and Y… Continue reading
In Parts 1, 2 and 3, we looked at statistical measures of datasets that were purely one-dimensional in nature, i.e., all the observations constituting the dataset were of the same type. Examples of such datasets are the time elapsed between eruptions of the Old Faithful Geyser
This short blog is in response to a recent support ticket querying whether particular domains display significant differences in terms of the ratio of referring subnets to referring IP addresses. We were asked what a “normal” value of this ratio could be.
In Parts 1 and 2, we have looked at the definitions of correlation, population and sample size. In this part, we will go through the essentials of the mathematics involved. Before we can proceed any further, however, we have to define certain fundamental statistical concepts.
Statistical Measures of a Data Set
In this section, it will be assumed that… Continue reading
(This is a guest post by Barrie Smith. His observations and insights are his own.)
I have recently been keeping a tab on Google page 1 results for “Payday Loans” – a search term historically showing results of websites ranking having used spammy, dodgy and/or underhand link building techniques.
Even in the year 2013 – more than 12 months after the Google Penguin algorithm update was… Continue reading