Detecting nonlinearity in time series by model selection criteria

This article analyzes the use of model selection criteria for detecting nonlinearity in the residuals of a linear model. Model selection criteria are applied for finding the order of the best autoregressive model fitted to the squared residuals of the linear model. If the order selected is not zero, this is considered as an indication of nonlinear behavior. The BIC and AIC criteria are compared to some popular nonlinearity tests in three Monte Carlo experiments. We conclude that the BIC model selection criterion seems to offer a promising tool for detecting nonlinearity in time series. An example is shown to illustrate the performance of the tests considered and the relationship between nonlinearity and structural changes in time series.     

Alternative Demand for University Outreach Services from Georgia's Golf Industry

Recent technological advancements have influenced farmers' growing demand for private information suppliers, a service traditionally monopolized by public outreach programs of land grant educational institutions. This waning appeal for university extension services creates pressure for these institutions to explore other market segments to broaden their audience coverage. This study demonstrates a potential market niche in Georgia's golf industry for university outreach services. Binomial logistic regression results indicate that significant demand ( a ) is negatively influenced by golf facility size; ( b ) is directly related to the level of industry experience of the golf superintendents; and ( c ) could come from facilities that require assistance in growth-, pest-, and employment-related concerns.     

Inter-firm Collaboration and Learning: The Case of the Japanese Automobile Industry

This paper investigates the effect of intra-firm Research and Development expenditure and inter-firm collaboration on firm performance. It is proposed that R&D expenditures enable a company to improve knowledge and assimilate the exchange of information within non-routine collaboration activities. The hypotheses are tested by an empirical study on inter-firm technical collaboration in Japanese Automobile firms during the years 1975–1995, using the method of first-order auto-regression on a panel data set. The findings indicate that R&D expenditure increases performance, but do not allow clear conclusions on the effects of collaborations.     

Measuring the Impact of Corporate Investment Announcements on Share Prices: The Spanish Experience

We bring together three disparate strands of literature to develop a comprehensive empirical framework to examine the efficiency of security analysts' earnings forecasts in Singapore. We focus specifically on how the increased uncertainty and the negative market sentiment during the period of the Asian crisis affected the quality of earnings forecasts. While we find no evidence of inefficiencies in the pre-crisis period, our results suggest that after the onset of the crisis, analysts (1) issued forecasts that were systematically upward biased; (2) did not fully incorporate the (negative) earnings-related news; and (3) predicted earnings changes which proved too extreme.     

Do stable strategic time periods exist? Towards new methodological and theoretical insights

This paper investigates a fundamental issue in the current research on strategic groups: the existence or non‐existence of the so‐called ‘stable strategic time periods’ (SSTPs). Our study provides new evidence by adding new methodological and theoretical insights. The research setting is the Spanish banking industry over a 15‐year period, 1983–1997. Unlike all prior longitudinal research that found SSTPs, the multi‐method procedure that we used in this study (i.e. equality of variance and covariance matrix and mean vector of strategic variables and a subsequent grouping analysis performed through the MCLUST) has led us to reject the existence of SSTPs in the industry under study. Based on these original findings, we conclude by suggesting a proposition which should be corroborated in future empirical studies on strategic groups. Copyright © 2007 John Wiley & Sons, Ltd.     

Homogeneity tests in meta-analysis: a Monte Carlo comparison of statistical power and Type I error

The statistical power and Type I error rate of several homogeneity tests, usually applied in meta-analysis, are compared using Monte Carlo simulation: (1) The chi-square test applied to standardized mean differences, correlation coefficients, and Fisher's r-to-Z transformations, and (2) S&H-75 (and 90 percent) procedure applied to standardized mean differences and correlation coefficients. Chi-square tests adjusted correctly Type I error rates to the nominal significance level while the S&H procedures showed higher rates; consequently, the S&H procedures presented greater statistical power. In all conditions, the statistical power was very low, particularly when the sample had few studies, small sample sizes, and presented short differences between the parametric effect sizes. Finally, the criteria for selecting homogeneity tests are discussed.     

Power in Profit-Maximizing Organizations

I study how profit-maximizing organizations make decisions. Members of organizations tend to have incompatible preferences over decisions, but willingness to pay for decisions plays a very limited role in actual decision making. A sizable empirical literature documents that people who provide critical services, are hard to replace, or deal effectively with external shocks are powerful; they have disproportionate influence over decisions. This can be profit maximizing because the right to shape the firm through its decisions renders the firm more attractive as an employer. Thus, the relative costliness of employees' departures should affect their relative power.