| 31 January 2017 |
This blog post is updated semi-regularly. It was last updated on the 22nd of June 2017.
There are quite a few features that we are working on and plan to go live with over the next few months. However, our months are measured in IT time (i.e., we will use our best endeavors, but things may go wrong and the timings may blow out).
Recap: the last 5 months
Since going into beta release in February, we have fixed many bugs, improved performance, and added lots of new features.
- Automatic updating of reports and analyses when the underlying data is changed.
- Various tools for automatically updating data from URLs or via R Code.
- Tools for manipulating data sets (Data Manipulation > Data Values).
- Lots of new analysis, visualization, and online reporting features (listed in the next three sub-sections).
You can read about each of these by searching this blog for posts on the various topics.
- Ability to insert raw data on any page (Insert > More (Analysis) > Tables > Raw Data)
- ANOVA and MANOVA (Insert > More (Analysis) > Analysis of Variance)
- Linear Discriminant Analysis (Insert > More (Analysis) > Machine Learning > Linear Discriminant Analysis).
- Gradient boosting (Insert > More (Analysis) > Machine Learning > Gradient Boosting).
- Support Vector Machines (Insert > More (Analysis) > Machine Learning > Support Vector Machine).
- Relative Importance Analysis, using generalizations of Johnson’s Weight are available for all the regression models (Insert > Regression (Analysis), setting the Output to Relative Importance Analysis).
- Better tools for missing data, including plots and Little’s MCAR test (Insert > More (Analysis) > Missing Data).
- A new sub-menu of MaxDiff tools: experimental designs, latent class analysis, and varying coefficients models (Insert > More (Analysis) > Marketing > MaxDiff).
- Labeled bubble charts (Insert > Visualization > Labeled Bubbleplot).
- Radar charts (Insert > Visualization > Radar Chart).
- Pictographs (Insert > Visualization > Pictograph).
- Scatterplot matrices (Insert > More (Analysis) > Correlation > Scatterplot Matrix).
- Sankey diagrams for visualizing missing data (Insert > Visualization > Sankey Diagram).
- Heatmaps for correlations matrices (Insert > More (Analysis) > Correlation > Correlation Matrix)
- Multidimensional Scaling (Insert > More (Analysis) > Dimension Reduction > Multidimensional Scaling (MDS))
- Multiple Correspondence Analysis (Insert > More (Analysis) > Dimension Reduction > Multiple Correspondence Analysis)
- Correspondence Analysis of a Square Table (Insert > More (Analysis) > Dimension Reduction > Correspondence Analysis of a Square Table)
- t-SNE(Insert > More (Analysis) > Dimension Reduction > t-SNE)
- Principal Components Analysis Biplot (Insert > More (Analysis) > Dimension Reduction > Principal Components Analysis Biplot
Online dashboards can be created using Export > Web Page (Document). Over the last few months we have added:
- Ability to hide specific analysis and pages, so that they appear when editing a report, but are invisible to users of the report.
- Adding controls that allow users to provide inputs (e.g., filters, specifying particular data to focus on).
- The ability to add hyperlinks between pages and to other documents, to aid in navigation (Insert > Hyperlink (Links)).
- Hiding outputs and pages from dashboards.
- Greater control over colors.
- Uploading of QPacks (i.e., data analysis performed using Q).
- Client logins and groups, restricting different users to different documents and to different sections within a document.
- Customization via CSS.
Bug fixing and general usability enhancement
Displayr is currently in beta release. You will find some irritating bugs. Some aspects of the user experience are clunky. At times the app is a bit too slow. We are working hard on these issues. This is our main day-to-day focus. We have lots of testing processes. We do user experience interviews every week. And, most importantly, we use Displayr ourselves every day.
Adding better support for dashboards
Over the next few months we will be rolling out the following features:
- Adding filtering. This already works in Displayr itself, we just need to hook it up so that it works when somebody is viewing a report that has been exported as a web page.
- An explore mode, so that users of dashboards can do their own custom analyses.
- Improved exporting.
- Tidying up the API.
Integrating charting and visualizations
Currently, charting and visualization work a bit differently. We will be integrating them and, at the same time, making them more flexible in terms of the data that they support.
A search tool to search through your data, your reports, and our menus and documentation.
In the longer term, our focus will be more on speed and larger data sets.
What have we missed? What is it that you want us to focus on? Please send me an email with any thoughts or feedback that you have (firstname.lastname@example.org).
Author: Tim Bock
Tim Bock is the founder of Displayr. Tim is a data scientist, who has consulted, published academic papers, and won awards, for problems/techniques as diverse as neural networks, mixture models, data fusion, market segmentation, IPO pricing, small sample research, and data visualization. He has conducted data science projects for numerous companies, including Pfizer, Coca Cola, ACNielsen, KFC, Weight Watchers, Unilever, and Nestle. He is also the founder of Q www.qresearchsoftware.com, a data science product designed for survey research, which is used by all the world’s seven largest market research consultancies. He studied econometrics, maths, and marketing, and has a University Medal and PhD from the University of New South Wales (Australia’s leading research university), where he was an adjunct member of staff for 15 years.