A play-the-winner-type urn design with reduced variability

We propose a new adaptive allocation rule, the drop-the-loser, that randomizes subjects in the course of a trial comparing treatments with dichotomous outcomes. The rule tends to assign more patients to better treatments with the same limiting proportion as the randomized play-the-winner rule. The new design has significantly less variable allocation proportion than the randomized play-the-winner rule. Decrease in variability translates into a gain in statistical power. For some values of success probabilities the drop-the-loser rule has a double advantage over conventional equal allocation in that it has better power and assigns more subjects to the better treatment. Acknowledgments. I thank Stephen Durham, the associate editor, and the referees for their helpful suggestions.     

Simulating the Evolution of the Implied Distribution

Motivated by the implied stochastic volatility literature (Britten–Jones and Neuberger, forthcoming; Derman and Kani, 1997; Ledoit and Santa–Clara, 1998) this paper proposes a new and general method for constructing smile–consistent stochastic volatility models. The method is developed by recognising that option pricing and hedging can be accomplished via the simulation of the implied risk neutral distribution. We devise an algorithm for the simulation of the implied distribution, when the first two moments change over time. The algorithm can be implemented easily, and it is based on an economic interpretation of the concept of mixture of distributions. It can also be generalised to cases where more complicated forms for the mixture are assumed.     

Factors affecting households’ seafood purchasing decisions in Auckland,New Zealand: an empirical analysis

This paper uses the binary choice model to identify the factors that are significantly influencing the household purchase decisions of seafood products for home consumption in Auckland, New Zealand. It is found that ‘quality’ and ‘cooking easiness’ are the main product attributes that significantly influence households’ choices of seafood in Auckland. Also, the representative household has shown a strong preference for fresh and other alternative seafood products, including processed, smoked and canned, over frozen products. Retail outlets are found to be more attractive to the household purchasing seafood for home consumption. The New Zealand seafood industry may find this baseline study useful as a guide to developing future research structure on the domestic market.     

What Prompts the People＇s Bank of China to Change Its Monetary Policy Stance？ Evidence from a Discrete Choice Model

In the present paper, we model the policy stance of the People＇s Bank of China （PBC） as a latent variable, and the discrete changes in the reserve requirement ratio, policy interest rates, and the scale of open market operations are taken as signals of movement of this latent variable. We run a discrete choice regression that relates these observed indicators of policy stance to major trends of macroeconomic and financial developments, which are represented by common factors extracted from a large number of variables. The predicted value of the estimated model can then be interpreted as the implicit policy stance of the PBC. In a second step, we estimate how much of the variation in the PBC＇ s implicit stance can be explained by measures of its policy objectives on inflation, growth and financial stability. We find that deviations of CPI inflation from an implicit target and deviations of broad money growth from the announced targets, but not output gaps, figure significantly in the PBC＇s policy changes.     

Statistics for Studies of Human Welfare

Developments in the recent past have substantially increased our ability to measure, compute, and communicate. We take the view that a corresponding improved understanding of processes in the life sciences will come about only through more intensive studies of properties of statistical methods and algorithms and transparent, open source computing environments.     

QUADRATIC TERM STRUCTURE MODELS FOR RISK-FREE AND DEFAULTABLE RATES

In this paper, quadratic term structure models (QTSMs) are analyzed and characterized in a general Markovian setting. The primary motivation for this work is to find a useful extension of the traditional QTSM, which is based on an Ornstein–Uhlenbeck (OU) state process, while maintaining the analytical tractability of the model. To accomplish this, the class of quadratic processes, consisting of those Markov state processes that yield QTSM, is introduced. The main result states that OU processes are the only conservative quadratic processes. In general, however, a quadratic potential can be added to allow QTSMs to model default risk. It is further shown that the exponent functions that are inherent in the definition of the quadratic property can be determined by a system of Riccati equations with a unique admissible parameter set. The implications of these results for modeling the term structure of risk-free and defaultable rates are discussed.     

FACTORS THAT AFFECT THE LONG‐TERM SURVIVAL OF MICRO,SMALL AND MEDIUM ENTERPRISES IN ETHIOPIA

There is an increasing recognition of the potential importance of micro, small and medium enterprises (MSMEs) for employment, income and poverty reduction in Ethiopia. Although the MSME sector contributes significantly to the national economy, the high failure rate among well established small businesses and enterprises is a matter of major concern. A random sample of 500 small businesses and enterprises selected from five major cities in Ethiopia were followed up for 6 years in order to assess the impact of influential factors that affect the long‐term survival and viability of small enterprises. Hazard ratios estimated from the Cox Proportional Hazards Model were used to quantify the impact of key predictors of survival. Businesses that ceased operation were characterised by inadequate finance (61%), low level of education (55%), poor managerial skills (54%), shortage of technical skills (49%), and inability to convert part of their profit to investment (46%). The study shows that participation in social capital and networking (iqqub schemes) was critically helpful for long‐term survival. Businesses that did not participate in iqqub schemes regularly were found to be 3.25 times more likely to fail in comparison with businesses that did. Results of the study have implications that could be used for developing policy initiatives that are aimed at fostering long‐term growth in small businesses and enterprises.     

The negative trade-off between risk and incentives: Evidence from the american whaling industry

This paper analyzes the trade-off between risk and incentives in the share contracts of the American whaling industry. Using a newly collected panel of 5378 individuals who sailed on more than 1000 whaling voyages from 1855–68, the response of sailors’ compensation to an increase in risk is estimated. The risks used to identify this response resulted from the commerce-raiding naval vessels of the Confederacy during the Civil War. As the Confederate cruisers sailed primarily in the Atlantic, and therefore posed far less of a threat to whaling voyages to other oceans, a quasi-experimental approach focusing on the differences between Atlantic voyages compared to others is implemented. The results are consistent with a negative trade-off between risk and incentives in the industry’s contracts. Moreover, evidence is found of selection among less risk-averse sailors and merchants into riskier voyages during the war.     

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.