An exploration that could have been performed very quickly in a simple spreadsheet, but was done (still pretty quickly) with R instead. All it does is explores how far off the shortcut calculation for estimating a Fahrenheit temperature from a Celsius temperature is by simply doubling the Celsius temperature and adding 20 degrees.
This example is an exploration of using Holt-Winters forecasting for anomaly detection. Mainly, it's geared towards an explanation of time-series decomposition and how that technique can be used to identify whether some set of data falls outside of an expected range.
This is a basic example of pulling data from the Adobe Analytics v2.0 API using R. It supports pulling just a single dimension and a single metric.
An example of doing some light text mining on a Twitter user's followers based on their profile descriptions
An example of doing some text mining of Google Analytics on-site search data.
An example of comparing pages on a scatterplot by assessing two different metrics for each page at once.
An example of identifying tortoise vs. hare content on the site by not getting too enamored with how much traffic a page garners immediately following its launch.
Playing around with the Twitter API and the Google Maps Geocoding API. And R.
An illustration of a data ecosystem by way of an interactive network diagram built from a Google Sheets document.
A light analysis of Twitter followers using data from the Twitter API and the Google Maps Geocoding API.
An exercise in matching a bar chart built in Domo and then "made over" by data visualization guru Lea Pica (in Domo). This document builds the same chart using ggplot2 in R.
This is a simple little head-to-head test on a couple of sites to see the differences in the speeds with which Google Analytics and Adobe Analytics return data through their respective APIs.
This report compares key metrics from Adobe Analytics to the comparable metrics from Google Analytics for a single site using data pulled using the RSiteCatalyst and googleAnalyticsR packages. For each metric, a boxplot shows the median, quartiles, and outliers of the daily difference between Adobe Analytics and Google Analytics for the metric. A line chart shows the actual values reported for the metric from both platforms. And, another line chart shows the % difference between the results for each day (with the Adobe Analytics value as the basis).
A slideshow build with RMarkdown as part of a session at eMetrics Berlin in November 2016. The intent was to provide a glimpse into the code required to access data through the Google Analytics API (using the googleAnalyticsR package) and then do some basic manipulation and visualization of the data.