New treemap features, as demonstrated by who's been buying lately
Written by Jason on October 31, 2011Following up on our recent mapping enhancements announcement, today we're going to review some of the new heatmap (aka treemap) features which can significantly enhance your visualization.
The big enhancements are:
- Depth support - trees can be up to four levels deep
- Custom messages - the box at the bottom can be used to provide additional information, and can be different for each element at each level
Let's take a quick look at an example where you can:
- Hover over an area to learn more about the selected item in the message box underneath the chart
- Left-click on a rectangle to descend a level and get a breakdown
- Right click to move up a level
This example summarizes certain insider trading activity since the recent market low in early October. Specifically, it shows all of the transactions that meet the following criteria:
- Transactions and filings that occurred between October 3rd and 28th, inclusive
- Officer transactions only - directors and other insiders are excluded
- The companies have officers buying at least $100k in aggregate over that period (so two $50k purchases by different people on different days would qualify the company, and then any officer transactions for that company - purchase or sale - would register)
We've implemented some custom sizing for the rectangles in this example - it's logarithmic and also takes into account the number of insiders and the number of companies in the industry, and weights purchases more heavily than sales. So you'll see that while large purchases (like the two officers at Opko Health in the surgical and medical instruments industry) remain prominent, there are three different companies in the state commercial banks industry that met our $100k officer purchase criteria, and is larger on the industry view.
Of course, you can scale items any way you choose through your formulas. You could scale them linearly by dollar amounts, number of participants, weighted by number of days from Oct 28th, or almost any other permutation you desire.
We set the color of the rectangle as the net number of officers that made purchases in the timeframe (number that made purchases less number that made sales). The more people that bought, the more green the rectangle at the industry and company level.
Hopefully this analysis is useful to you, and helps you understand what you can do with Tuhunga.
