DataLab is a compact statistics package aimed at exploratory data analysis. Please visit the DataLab Web site for more information....



Introduction

The process of data analysis is usually a two-step process. First, the scientist has to explore the inner structure of the data in order to get an idea on how to further process it. Secondly, a specific model has to be set up which provides a satisfying relationship between measured and target variables. DataLab certainly puts emphasis on the first aspect of data analysis; it could be classified as a program for exploratory data analysis.

DataLab is quite different from other statistical packages in that it does not claim to cover all statistical techniques. Moreover, DataLab has been deliberately designed to be not a full-featured, blown-up package but rather to be a handy and easy-to-use tool in everyday work with experimental data. It provides only some of the more fundamental methods and the selection of these methods has certainly been guided by a personal need of the author. Therefore DataLab provides tools for both the visualization and interpretation of data.

The system is based on a powerful but easy-to-use graphical user interface. DataLab provides a collection of important methods to edit, visualize and interpret numeric data.

These are some of the features of DataLab:

  • various methods to import and export data
  • built-in numeric data editor
  • a large number of chart types: point plots, line plots, spectra, x-y plots, contour lines, histograms, box plots, table survey, scatter plots, normal probability plots etc.
  • on-screen 3D-rotation of data
  • color maps to visualize three-dimensional relationships
  • time series visualization and analysis
  • built-in math calculator allows to process measured data
  • statistical tests (t-test, F-test, Kolmogorov-Smirnov, Dean-Dixon, correlation test, etc.)
  • mathematical/statistical functions: linear regression (line, polynomial, hyperbola, etc.), rank correlation, multiple linear regression, principal component analysis, neural networks, KNN-modelling, hierarchical clustering (dendrograms), random number generator, integration of signals, Fourier transformation, filter, smoothing
  • automatic feature selection: forward selection (MLR), step wise regression, growing neural networks
  • displayed windows are freely configurable in size and colors
  • zoom and pan of data
  • data can be labeled by additional class information
  • the axes and the titles of the graphs as well as single data items can be labeled almost arbitrarily
  • sorting and mixing of data
  • matrix clipboard allows flexible reorganisation of data
  • curve digitizer allows to convert printed measurement curves into digital data
  • automatic update


Last Update: 2012-Jul-25