The Multilevel Approach to Repeated Measures for Complete and Incomplete Data

Repeated measurements often are analyzed by multivariate analysis of variance (MANOVA). An alternative approach is provided by multilevel analysis, also called the hierarchical linear model (HLM), which makes use of random coefficient models. This paper is a tutorial which indicates that the HLM can be specified in many different ways, corresponding to different sets of assumptions about the covariance matrix of the repeated measurements. The possible assumptions range from the very restrictive compound symmetry model to the unrestricted multivariate model. Thus, the HLM can be used to steer a useful middle road between the two traditional methods for analyzing repeated measurements. Another important advantage of the multilevel approach to analyzing repeated measures is the fact that it can be easily used also if the data are incomplete. Thus it provides a way to achieve a fully multivariate analysis of repeated measures with incomplete data. This revised version was published online in June 2006 with corrections to the Cover Date.     

An Analysis of Differences in Earnings Between Small and Large Commercial Banks

This paper introduces a simple, yet rich, measure of efficiency changes based on the revenue-generating-ability (RGA) principle. Using this principle, we explain the connections between efficiency changes and the variables, such as pretax profits, interest expense, non-interest expense, profit margins, loan loss provision, and asset quality. These connections are used to explain earnings differences between small and large commercial banks.     

Spreads, depths, and the impact of earnings information: an intraday analysis

For a sample of NYSE firms, we show that wide spreads are accompaniedby low depths, and that spreads widen and depths fall in responseto bigger volume. Spreads widen and depths fall in anticipationof earnings announcements; these effects are more pronouncedfor announcements with larger subsequent price changes. Spreadsare also wider following earnings announcements, but this effectdissipates quickly after controlling for volume. Collectively,our results suggest liquidity providers are sensitive to changesin information asymmetry risk and use both spreads and depthsto actively manage this risk.     

Innovative productivity and returns to scale in the pharmaceutical industry

This study shows that firms in the pharmaceutical industry experience decreasing returns to scale in R & D as the level of R & D expenditures rises. The paper presents the results of our study of the innovative output of 16 pharmaceutical firms over a 19 year period. Given the strong correlation between R & D budgets and firm size, our study suggests the wave of mergers in the industry may yield less innovative productivity than managers expect.     

Sustaining Collusion in Growing Markets

The impact of demand growth on the collusion possibilities is investigated in a Cournot supergame where market growth may trigger future entry and the collusive agreement is enforced by the most profitable ‘grim trigger strategies’ available. It is shown that even in situations where perfect collusion can be sustained after entry, coping with a potential entrant in a market which is growing over time may completely undermine any pre‐entry collusive plans of the incumbent firms. This is because, before entry, a deviation and the following punishment phase may become more attractive thanks to their additional effect in terms of delaying entry.     

The Effects of Screening and Monitoring on Credit Rationing of SMEs

In this paper, we seek to empirically assess which determinants of the capability and incentives of banks to screen and monitor firms are significant in explaining credit rationing to Italian SMEs. After testing for the presence of non‐random selection bias and the potential endogeneity of some determinants of interest, the probit model results we obtain suggest that the average banking size and the multiple banking relationship phenomenon are statistically significant factors affecting credit rationing, presumably through their impact on the aforementioned banks' capability and incentives. Other potential determinants of banks' incentives to monitor and screen, such as local banking competition and firm' capacity to collateralize, are never significant. However, when we split the sample according to the level of competition in credit markets, we find that the estimated marginal effects of all significant determinants of interest are larger in absolute value than those obtained when using the whole sample.     

Intertemporal substitution and the liquidity effect in a sticky price model

The liquidity effect, defined as a decrease in nominal interest rates in response to a monetary expansion, is a major stylized fact of the business cycle. This paper first confirms that, with separable preferences, a low degree of intertemporal substitution in consumption is a necessary condition for the existence of the liquidity effect. In contrast to this result, in a model with non-separable preferences and capital accumulation it takes an implausibly high elasticity of intertemporal substitution to produce a liquidity effect. The robustness of these results to alternative degrees of nominal rigidities, capital adjustment costs and stochastic monetary processes is also analysed. We conclude that price stickiness, by itself, does not guarantee the existence of a liquidity effect.     

Bayesian MCMC Mapping of Quantitative Trait Loci in a Half-sib Design: a Graphical Model Perspective

Graphical models provide a powerful and flexible approach to the analysis of complex problems in genetics. While task-specific software may be extremely efficient for any particular analysis, it is often difficult to adapt to new computational challenges. By viewing these genetic applications in a more general framework, many problems can be handled by essentially the same software. This is advantageous in an area where fast methodological development is essential. Once a method has been fully developed and tested, problem-specific software may then be required. The aim of this paper is to illustrate the potential use of a graphical model approach to genetic analyses by taking a very simple and well-understood problem by way of example.