1352.0.55.139 - Research Paper: Discrete Choice Panel Data Modelling Using the ABS Business Longitudinal Database (Methodology Advisory Committee), Nov 2013  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 21/02/2014  First Issue
   Page tools: Print Print Page Print all pages in this productPrint All
  • About this Release

This study explores a number of discrete choice panel data models to analyse the effects of three factors on innovation: flexible working arrangements, information technology, and collaboration, while controlling for the effects of other important variables, such as industry, business size, competition, and market location. The study also examines whether firms that innovated in the past are more likely to innovate in subsequent periods.

The econometric models examined range from the relatively simple pooled model to the more complex dynamic probit model. The aim is to assess both the static and dynamic relationships and to take care of the categorical dependent variable and the longitudinal nature of the dataset.


The analyses undertaken are in the context of the Australian small and medium-sized firms. Three waves of the ABS Business Longitudinal Database are used: 2007–2008, 2008–2009, and 2009–2010.