A framework that brings together an abundance of common statistical models found across packages into a unified interface, and provides a common architecture for estimation and interpretation, as well as bridging functions to absorb increasingly more models into the collective library. Zelig allows each individual package, for each statistical model, to be accessed by a common uniformly structured call and set of arguments. Moreover, Zelig automates all the surrounding building blocks of a statistical work-flow--procedures and algorithms that may be essential to one user's application but which the original package developer did not use in their own research and might not themselves support. These include bootstrapping, jackknifing, and re-weighting of data. In particular, Zelig automatically generates predicted and simulated quantities of interest (such as relative risk ratios, average treatment effects, first differences and predicted and expected values) to interpret and visualize complex models.
The Zelig core team is pleased to announce the alpha release of Zelig 4.
Designated as the "Developer Update", Zelig 4 offers a wide-range of improvements to ease the process of adding new statistical models to the already extensive Zelig software suite. Significantly, this release is packaged with a brand-new API, geared towards reducing the complexity and length of Zelig's development functions - the zelig2, param and qi methods. In addition to this, Zelig now brandishes a package-creator (zelig.skeleton) that operates in the same vein as R's core function "package.skeleton".
In addition to changes in the development toolkit, Zelig has now been split across 13 distinct packages. This change has been made to refine the scope of Zelig and its add-ons. In particular, this restructuring of Zelig into a full software suite allows developers to contribute, develop and repair add-on packages without tinkering with the Zelig API and core functionality.
While this release's prime focus has been improving the developer toolkit and restructuring the software suite, Zelig 4 offers an end-user experience completely identical to previous versions. That is, zelig's basic functions - zelig, setx and sim - ostensibly remain unchanged in functionality for available statistical models.
For full details concerning changes between Zelig 3.5 and Zelig 4, please refer to: http://zeligdev.github.com/
Some of the new available features are:
A revised developer API. The primary developer methods - zelig2, param and sim - have been reimplemented to use a sleeker, simpler API. For information, please read the Zelig developer's manual found here: http://zeligdev.github.com/files/booklet.pdf
The core package has been restructured and minimized. In particular, Zelig core now contains only code essential to its operation, while all non-essential tasks have been made into specific R-packages. For a complete list of official Zelig packages, please refer to: https://github.com/zeligdev
Development tools for contributors have been added to the core package. In particular, the "zelig.skeleton" function is packaged within Zelig-core in order to facilitate the rapid development of new Zelig packages.
The Zelig software suite has grown to include a total of 7 R-packages. This change offers a simple and easy method for ensuring that development and bug-fixing within any particular Zelig add-on will leave the remainder of the Zelig software suite unchanged.
A hook API has been integrated into the core package, in order to reduce the necessity to directly alter the zelig, setx and sim methods.
Roxygen-compliant documentation has become standard in all Zelig packages. This offers an easy way to manage Rd documentation, dependencies and exports from within the R code itself. That is, documentation is more tightly paired with the actual R code. For more information about Roxygen, please refer to: http://roxygen.org/
Zelig is now on GitHub! Fork an add-on package or contribute bug-finds today!
For a full listing of official packages and their repositories, please see: https://github.com/zeligdev
The following comprises a list of relevant information for Zelig 4:
For any particular questions on developing new Zelig models, please send all mail to: firstname.lastname@example.org
Zelig v4.0-4 Release Notes (May 16, 2011)
This document is a brief overview of the current state of the Zelig project as of the 4.0-3 release. This release hopes to maintain the canonical Zelig syntax and interface for end-users, while supplying developers with tools to aid in the development of effective statistical modeling techniques. Emphasis has been placed on readability and modularity.
As a result of this gargantuan change, a plethora of features, API functionality, and documentation has been added to the Zelig R-package. Several previously existing models, however, have been removed temporarily or moved from the Zelig core package to more-specific Zelig extensions.
The Zelig software suite is an easy-to-use R-package geared towards making complex statistical techniques available to end users, particularly those researching the quantitative social sciences. In particular, it offers unifying syntax and programming-style between seemingly disparate and unrelated statistical mdoels.
To facilitate this purpose, Zelig (as of May 16th, 2011) includes an array of programming tools, geared towards allowing the rapid development, debugging, and inclusion of new statistical models. That is, Zelig now facilitates and encourages collaboration between novel and pre-existing statistical packages.
Zelig is a collaborative effort by Harvard's Institute for Quantitive Social Sciences (Harvard IQSS). Publications, software releases, and additional information can be found at: http://gking.harvard.edu/ http://iq.harvard.edu/
Zelig is licensed under the GNU General Public License version 2, and as such can be freely used and edited given proper attribution to Harvard's IQSS Department.
This release offers a large variety of coding style changes, as well as core functionality. Please carefully read the following: