Ireland's Innovation Performance: 1991-2005

In this paper we use data from the five waves of the Irish Innovation Panel (IIP) to profile the innovation performance of manufacturing plants in Ireland and Northern Ireland over the period 1991 to 2005. Despite considerable public sector investment on both sides of the border levels of innovation activity have remained broadly similar throughout this period although somewhat different trends are evident in Ireland and Northern Ireland. In terms of product innovation for example, the proportion of manufacturing plants making product changes has increased 5 per cent in Ireland and just over 7 per cent in Northern Ireland. In terms of process innovation a decline of almost 7 per cent in Ireland has been accompanied by a 7 per cent increase in Northern Ireland. These trends provide some evidence of convergence in innovation performance over the 1991 to 2005 period. This is evident in the narrowing gap between the proportion of product innovators in Ireland and Northern Ireland, convergence in the proportion of plants undertaking process innovation and in terms of the increasingly similar proportions of sales derived from innovative products. Looking in more detail at the determinants of manufacturing innovation emphasises the importance of R&D and backwards supply chain linkages as sources of new knowledge for innovation. Other external linkages prove less important suggesting the value of policy initiatives designed to promote knowledge sharing. We also find a significant negative innovation effect from legislative restrictions on plants’ product innovation. Public support for both product and process innovation are having positive effects on innovation outputs at the level of the individual plant. Future research interest centres on the contrast between this strong positive result at firm level and the more modest increases in innovation among the population of firms in Ireland and Northern Ireland.

Service Innovation, Embeddedness and Business Performance: Evidence from Northern Ireland

This paper explores the factors that determine innovation by service firms, and in particular the contribution of intra- and extra-regional connectivity. Subsequently, it is examined how service firms' innovation activity relates to productivity and export behaviour. The empirical analysis is based on matched data from the 2005 UK Innovation Survey - the UK component of the 4th Community Innovation Survey (CIS) - and the Annual Business Inquiry for Northern Ireland. Evidence is found of negative intra-regional embeddedness effects, but there is a positive contribution to innovation from extra-regional connectivity, particularly links to customers. Relationships between innovation, exporting, and productivity prove complex, but suggest that innovation itself is not sufficient to generate productivity improvements. Only when innovation is combined with increased export activity are productivity gains evident.

Technological regimes, Schumpeterian patterns of innovation and firm-level productivity growth

The article investigates the relationships between technological regimes and firm-level productivity performance, and it explores how such a relationship differs in different Schumpeterian patterns of innovation. The analysis makes use of a rich dataset containing data on innovation and other economic characteristics of a large representative sample of Norwegian firms in manufacturing and service industries for the period 1998–2004. First, we decompose TFP growth into technical progress and efficiency changes by means of data envelopment analysis. We then estimate an empirical model that relates these two productivity components to the characteristics of technological regimes and a set of other firm-specific factors. The results indicate that: (i) TFP growth has mainly been achieved through technical progress, while technical efficiency has on average decreased; (ii) the characteristics of technological regimes are important determinants of firm-level productivity growth, but their impacts on technical progress are different from the effects on efficiency change; (iii) the estimated model works differently in the two Schumpeterian regimes. Technical progress has been more dynamic in Schumpeter Mark II industries, while efficiency change has been more important in Schumpeter Mark I markets.

Incremental learning from stream data

Recent years have witnessed an incredibly increasing interest in the topic of incremental learning. Unlike conventional machine learning situations, data flow targeted by incremental learning becomes available continuously over time. Accordingly, it is desirable to be able to abandon the traditional assumption of the availability of representative training data during the training period to develop decision boundaries. Under scenarios of continuous data flow, the challenge is how to transform the vast amount of stream raw data into information and knowledge representation, and accumulate experience over time to support future decision-making process. In this paper, we propose a general adaptive incremental learning framework named ADAIN that is capable of learning from continuous raw data, accumulating experience over time, and using such knowledge to improve future learning and prediction performance. Detailed system level architecture and design strategies are presented in this paper. Simulation results over several real-world data sets are used to validate the effectiveness of this method.