April 27, 2011
Data Wrangler is a web-based service from Stanford University’s Visualization Group. Similar to Google Refine it allows interactive cleansing and transformation of messy, real-world data for further analysis in e.g. Excel, R, Protovis etc.
To learn how to use Wrangler read this research paper or this blog post.
Wrangler is still in its alpha version so if you want to give feedback to the developers, just go for it.
Wrangler Demo Video from Stanford Visualization Group on Vimeo.
April 27, 2011
Welcome to the best network – Vodafone has a new commercial and I kind of like it 🙂
The song “Way back home” is scored by the australian duo Bag Raiders
March 16, 2011
How often do you have to deal with non-normal data? Do you know what to do with it? In his article “Dealing with Non-normal Data: Strategies and Tools” Arne Buthmann explains the common reasons for non-normal data and how to handle it.
Addressing Reasons for Non-normality
Reason 1: Extreme Values
Reason 2: Overlap of Two or More Processes
Reason 3: Insufficient Data Discrimination
Reason 4: Sorted Data
Reason 5: Values Close to Zero or a Natural Limit
Reason 6: Data Follows a Different Distribution
No Normality Required
He states that “Some statistical tools do not require normally distributed data. To help practitioners understand when and how these tools can be used, the table below shows a comparison of tools that do not require normal distribution with their normal-distribution equivalents.”
|Comparison of Statistical Analysis Tools for Normally and Non-Normally Distributed Data
|Tools for Normally Distributed Data
||Equivalent Tools for Non-Normally Distributed Data
||Mann-Whitney test; Mood’s median test; Kruskal-Wallis test
||Mood’s median test; Kruskal-Wallis test
||One-sample sign test
|F-test; Bartlett’s test
|Individuals control chart
||Weibull; log-normal; largest extreme value; Poisson; exponential; binomial
March 16, 2011
In the real world we will almost always come across missing values in data due to many reasons. This problem must be adressed to produce reliable statistical results. First of all we need to identify what is missing. Then ask yourself why the data is missing and what it means. After you have answered those questions you need to deal with the missing values. Those are the steps to take:
1. When missing values are few and lay far apart then do nothing.
2. When a column has a significant number of missing values then create a variable for missing, present values (0/1).
3. When a column has a significant number of missing values then replace the missing value with a constant value e.g. mean, median or mode.
4. When a column and its values are essential to producing accurate predictions then estimate the missing value based on other, non-missing data elements.
December 26, 2010
“I kid you not, statistics are now the sexiest subject on the planet.”
Hans Rosling is my idol, his online lectures are of this eye-opening, mind-expanding and funny kind. He is not only the developer of gapminder, a really cool information visualization software for animation of statistics but also an international known medical doctor, academic, statistician and public speaker.
In his videos ‘The Joy of Stats’ he shows exactly why some people (including me) have a passion for statistics, it is exciting and fun to find the story behind the data, tell it in a visual appealing way and make sense of the world.
You can find many more great lectures on TED or of The Joy of Stats series.
UPDATE: The Joy of Stats is now available in its entirety on Gapminder
November 19, 2010
On his blog Occam’s razor about web analytics Avinash Kaushik wrote a post with the topic “Beginner´s guide to web analytics: Ten steps to love & success”. Being an expert in web analytics he gives a practical introduction into this field. This is an overview about the ten steps that you should follow according to him. The whole post gives a useful outline how to get started with web analytics, so check out here.
Step 1: Visit the website. Note objectives, customer experience, suckiness.
Step 2: How good is the acquisition strategy? Traffic Sources Report.
Step 3: How strongly do Visitors orbit the website? Visitor Loyalty & Recency.
Step 4: What can I find that is broken and quickly fixable? Top Landing Pages.
Step 5: What content makes us most money? $Index Value Metric.
Step 6: How Sophisticated Is Their Search Strategy? Keyword Tag Clouds.
Step 7: Are they making money or making noise? Goals & Goal Values.
Step 8: Can the Marketing Budget be optimized? Campaign Conversions/Outcomes.
Step 9: Are we helping the already convinced buyers? Funnel Visualization.
Step 10: What are the unknown unknowns I am blind to? Analytics Intelligence.
Click here for full article.