Long-term Success of Adaptation Strategies: Evidence from Finnish Companies

This study analyses the success of adaptation strategies applied by Finnish companies during the recession in 1989–93. Five different strategies emerged from data gleaned from a questionnaire answered by 750 companies. In the medium term, investment in new product development and marketing and in the acquisition of new customers was the most successful strategy while a strategy heavily based on negotiating finance contracts and restructuring was the most unsuccessful. To evaluate the long-term effects of adopted strategies, financial statements from 644 companies in 1994 and 1997 were analysed. The results supported the medium-term conclusions. In general, they suggested that in the long-term, active adaptation strategies are most likely to lead to sustained success whereas passive (financial) strategies are associated with a high risk of failure.     

A further note on a theorem on the difference of the generalized inverses of two nonnegative definite matrices

Andrews and Phillips (1986) gave a simplified proof for the result that established the nonnegative definiteness of the difference of the Moore-Penrose inverses of two nonoegative definite matrices, a result originally due to Milliken and Akdeniz (1977), The purpose of this paper is to offer a simple proof for a generalization of this result,     

Regression analysis with two mixed forms of heteroscedasticity

Regression models are here considered in which disturbances are related to both the expectation of the dependent variable and a linear conbination of certain auxiliary variables. The maximum likelihood and weighted least squares estimators are compared in estimating the form of heteroscedasticity and regression coefficients. Also a test for heteroscedasticity is discussed. Finally an example is worked out for the purpose of illustration.     

Optimal Designs for Binary Logistic Regression with a Qualitative Classifier with Independent Levels

Dose response studies arise in many medical applications. Often, such studies are considered within the framework of binary-response experiments such as success-failure. In such cases, popular choices for modeling the probability of response are logistic or probit models. Design optimality has been well studied for the logistic model with a continuous covariate. A natural extension of the logistic model is to consider the presence of a qualitative classifier. In this work, we explore D-, A-, and E-optimal designs in a two-parameter, binary logistic regression model after introducing a binary, qualitative classifier with independent levels.     

Long-term Stability of Social Participation

The long-term stability of social participation was investigated in a representative urban population of 415 men and 579 women who had taken part in the nationwide Mini-Finland Health Survey in the years 1978–1980 and were re-examined 20 years later. Stability was assessed by means of the following tracking coefficients: kappa, proportion of stability, Spearman correlation, and stability coefficient (odds ratio) for engagement in clubs and voluntary societies, cultural and sports attendance, congregational activity, outdoor and productive activity, cultural interests, and hobby activity. The tracking results showed fair to moderate stability for the above measures of social participation, varying from the highest weighted kappa (0.43) for congregational activity in women to the lowest weighted kappa (0.22) for cultural and sports attendance in men. Migration from rural to urban community did not affect the stability of social participation. The underlying latent construct ‘social participation’ based on a two-wave multi-variable model showed good stability (r = 0.65). We conclude that social participation rests on a fairly stable foundation making it a useful tool for long-term epidemiological studies.     

Measurement and socio-demographic variation of social capital in a large population-based survey

Objectives The main objective of this study was to describe the variation of individual social capital according to socio-demographic factors, and to develop a suitable way to measure social capital for this purpose. The similarity of socio-demographic variation between the genders was also assessed. Data and methods The study applied cross-sectional data from the national Finnish Health 2000 survey (n = 8,028) which represents the adult population, aged 30 years and over. Several variables indicating social capital were condensed to dimensions on the basis of factor analysis. Participants were categorized into tertiles in each dimension of social capital by means of factor scores. The multinomial logistic regression model was used to produce the adjusted prevalences for the dimensions of social capital according to socio-demographic categories (age, gender, education, living arrangements, income, and type of region). Results Three dimensions of social capital were distinguished: social support, social participation and networks, and trust and reciprocity. Age had an inverse association with social support as well as participation and networks, and a curvilinear association between age and trust and reciprocity, the oldest age groups showing the highest level of trust. Married persons and those in the highest educational and income groups tended to have more social capital than other persons. Residents of urban and rural regions did not systematically differ from each other in their level of social capital although residents of urban regions participated less and showed less trust than people living in semi-urban or rural regions. Social support varied significantly with gender. The decline of social support by age was steeper in women than in men. Social participation and networks increased with education, the gradient appearing steeper among men. The difference between married and cohabiting men was substantial compared to women when it came to trust. Conclusions People who are young, married, educated, and well-off have plenty of social capital. This information might help various services to concentrate the actions on the people in danger of social exclusion. Our results also form a basis for the future by allowing the changes in social capital to be examined over time and over different studies.     

The Role of Profit Sharing in Dual Labour Markets with Flexible Outsourcing

We combine profit sharing for high‐skilled workers and outsourcing of low‐skilled tasks in a partly imperfect dual domestic labour market, which means that only low‐skilled labour is represented by a labour union. In that framework we analyse how the implementation of profit sharing for high‐skilled workers influences the amount of outsourcing and the labour market outcome for low‐skilled worker. By doing this, we use some specific assumptions, e.g. exponentially increasing outsourcing costs or the wage for low‐skilled workers will be determined by a union whereas the wage for high‐skilled workers is given. Assuming that low‐skilled labour and outsourcing are interchangeable we show that profit sharing has a positive effect on the wage for low‐skilled workers and helps to decrease wage dispersion. However, under these circumstances, profit sharing enhances outsourcing. Concerning the employment effects for high‐ and low‐skilled workers, we show that there is an employment reducing effect due to higher wages for low‐skilled work, which can be offset by higher productivity of highly skilled workers, as the domestic labour inputs complement each other.     

Let us do the twist again

Krämer (Sankhy $\bar{\mathrm{a }}$ 42:130–131, 1980) posed the following problem: “Which are the $\mathbf{y}$ , given $\mathbf{X}$ and $\mathbf{V}$ , such that OLS and Gauss–Markov are equal?”. In other words, the problem aimed at identifying those vectors $\mathbf{y}$ for which the ordinary least squares (OLS) and Gauss–Markov estimates of the parameter vector $\varvec{\beta }$ coincide under the general Gauss–Markov model $\mathbf{y} = \mathbf{X} \varvec{\beta } + \mathbf{u}$ . The problem was later called a “twist” to Kruskal’s Theorem, which provides conditions necessary and sufficient for the OLS and Gauss–Markov estimates of $\varvec{\beta }$ to be equal. The present paper focuses on a similar problem to the one posed by Krämer in the aforementioned paper. However, instead of the estimation of $\varvec{\beta }$ , we consider the estimation of the systematic part $\mathbf{X} \varvec{\beta }$ , which is a natural consequence of relaxing the assumption that $\mathbf{X}$ and $\mathbf{V}$ are of full (column) rank made by Krämer. Further results, dealing with the Euclidean distance between the best linear unbiased estimator (BLUE) and the ordinary least squares estimator (OLSE) of $\mathbf{X} \varvec{\beta }$ , as well as with an equality between BLUE and OLSE are also provided. The calculations are mostly based on a joint partitioned representation of a pair of orthogonal projectors.     

Should “open innovation” change innovation policy thinking in catching-up economies? Considerations for policy analyses

This paper provides a review of the current state of academic and policy-level debates on “open innovation” by elaborating on the relevance of the concept of open innovation for innovation policy-making in catching-up economies. The paper shows that paradoxes and contradictions exist between the “mainstream” innovation discourse and the development challenges of the catching-up countries that have lead to “de-contextualization” of the innovation policy discourse. The paper argues that applying the concept of open innovation in its ideal-type form to the catching-up context is likely to reinforce these de-contextualization tendencies. This problem can be remedied by more conscious attention to the basic contradictions and paradoxes, which requires a more comprehensive and systemic analytical focus on innovation and technological development at the levels of firm, industry and policy.     

A test of the mean square error criterion for shrinkage estimators

A test for choosing between a shrinkage estimator and the least squares estimator is described and a central-F approximation to the test statistic is considered. An example from the literature was analysed using the test procedure proposed here. The power of the test was studied by means of simulation.