An adaptive sampling scheme guided by BART—with an application to predict processor performance

The evaluation of new processor designs is an important issue in electrical and computer engineering. Architects use simulations to evaluate designs and to understand trade‐offs and interactions among design parameters. However, due to the lengthy simulation time and limited resources, it is often practically impossible to simulate a full factorial design space. Effective sampling methods and predictive models are required. In this paper, the authors propose an automated performance predictive approach which employs an adaptive sampling scheme that interactively works with the predictive model to select samples for simulation. These samples are then used to build Bayesian additive regression trees, which in turn are used to predict the whole design space. Both real data analysis and simulation studies show that the method is effective in that, though sampling at very few design points, it generates highly accurate predictions on the unsampled points. Furthermore, the proposed model provides quantitative interpretation tools with which investigators can efficiently tune design parameters in order to improve processor performance. The Canadian Journal of Statistics 38: 136–152; 2010 © 2010 Statistical Society of Canada     

Higher order moments of order statistics from INID symmetric random variables

In this paper we present analogues of Balakrishnan's (1989) relations that relate the triple and quadruple moments of order statistics from independent and nonidentically distributed (I.NI.D.) random variables from a symmetric distribution to those of the folded distribution. We then apply these results, along with the corresponding recurrence relations for the exponential distribution derived recently by Childs (2003), to study the robustness of the Winsorized variance.     

Asymptotic properties of a generalized regression-type predictor of a finite population variance in probability sampling

A system of predictors for estimating a finite population variance is defined and shown to be asymptotically design-unbiased (ADU) and asymptotically design-consistent (ADC) under probability sampling. An asymptotic mean squared error (MSE) of a generalized regression-type predictor, generated from the system, is obtained. The suggested predictor attains the minimum expected variance of any design-unbiased estimator when the superpopulation model is correct. The generalized regression-type predictor and the predictor suggested by Mukhopadhyay (1990) are compared.     

Optimal estimation of a finite population mean under generalized random permutation models

Employing certain generalized random permutation models and a general class of linear estimators of a finite population mean, it is shown that many of the conventional estimators are “optimal” in the sense of minimum average mean square error. Simple proofs are provided by using a well-known theorem on UMV estimation. The results also cover certain simple response error situations.     

‘Public toilets for a diverse neighborhood’: Spatial purification practices in community development discourse

This paper explores the dialectic of place and community identity in Mount Pleasant, a multi–ethnic and multi–class U.S. neighborhood where definitions of place are hotly contested among its residents. In a grant proposal for public toilets, Mount Pleasant writers use linguistic strategies such as presupposition, deixis, and contrast, coupled with discursive themes of filth and geography, to construct a core of the Mount Pleasant community. The writers place themselves and people who share their values in that core, and immigrants at the margins. These strategies serve as a discursive type of spatial purification practice (cf. Sibley 1988) through which the grantwriters set up a moral and spatial order where they and other core community members are deemed to use space ‘appropriately’, and thus inhabit positive moral positions, while immigrant community members’ imputed ‘inappropriate’ use of space is used to construct negative moral positions for them.     

Real option games with R&D and learning spillovers

We demonstrate to decision makers how to optimally make costly strategic pre-investment R&D decisions in the presence of spillover effects in an option pricing framework with analytic tractability. Decisions are modeled as impulse-type controls with random outcome. Two firms face two decisions that are solved interdependently in a two-stage game. The first-stage decision is: What is the optimal level of coordination (optimal policy/technology choice)? The second-stage decision is: What is the optimal effort for a given level of the spillover effects and the cost of information acquisition? The framework is extended to a two-period closed-loop stochastic game with (path-dependency inducing) switching costs that make strategy revisions harder. When conditions of learning-by-doing exist, we find that strategy shifts are easier to observe in market environments of high growth and high volatility.     

UNCERTAIN DEMAND,CONSUMER LOSS AVERSION,AND FLAT‐RATE TARIFFS

We consider a model of firm pricing and consumer choice, where consumers are loss averse and uncertain about their future demand. Possibly, consumers in our model prefer a flat rate to a measured tariff, even though this choice does not minimize their expected billing amount—a behavior in line with ample empirical evidence. We solve for the profit‐maximizing two‐part tariff, which is a flat rate if (a) marginal costs are not too high, (b) loss aversion is intense, and (c) there are strong variations in demand. Moreover, we analyze the optimal nonlinear tariff. This tariff has a large flat part when a flat rate is optimal among the class of two‐part tariffs.     

OUTSOURCING WHEN INVESTMENTS ARE SPECIFIC AND INTERRELATED

Using the universe of large Canadian manufacturing firms in 1988 and 1996, we investigate to what extent outsourcing patterns concord with the predictions of a simple property rights model. The unique availability of disaggregate information on outputs as well as inputs permits the construction of a detailed measure of vertical integration. We rely on five measures of technological intensity to proxy for investments that are likely to be specific to a buyer–seller relationship. A theoretical model that allows for varying degrees of investment specificity and interrelatedness—externalities between buyer and supplier investments—guides the analysis. Property rights predictions on the link between investment intensities and optimal ownership are strongly supported, but only for transactions with low interrelatedness. High specificity and low risk of appropriation strengthen the predictions in the model and in the data.     

Identifying Technology Spillovers and Product Market Rivalry

The impact of R&D on growth through spillovers has been a major topic of economic research over the last thirty years. A central problem in the literature is that firm performance is affected by two countervailing “spillovers” : a positive effect from technology (knowledge) spillovers and a negative business stealing effects from product market rivals. We develop a general framework incorporating these two types of spillovers and implement this model using measures of a firm's position in technology space and productmarket space. Using panel data on U.S. firms, we show that technology spillovers quantitatively dominate, so that the gross social returns to R&D are at least twice as high as the private returns. We identify the causal effect of R&D spillovers by using changes in federal and state tax incentives for R&D. We also find that smaller firms generate lower social returns to R&D because they operate more in technological niches. Finally, we detail the desirable properties of an ideal spillover measure and how existing approaches, including our new Mahalanobis measure, compare to these criteria.     

The moderating effects of home region orientation on R&D investment and international SME performance: Lessons from Korea

Drawing from the resource-based view, we investigate how firm and country combinations affect international SME performance. Using a sample of 2676 international Korean SMEs, we explore the relationships among SME R&D investment, home region orientation and financial performance. Results show that R&D investment has a horizontally inverted S-shaped relationship with performance reflecting cost leadership, stuck in the middle, and differentiation strategies. We also find that a home region orientation moderates the relationship between R&D investment and performance. Home region orientation positively moderates the relationship when R&D investment reflects cost leadership or differentiation but negatively moderates a stuck in the middle strategy.