Shulamith T. Gross

Does It Matter Where Countries Are? Proximity to Knowledge, Markets and Resources, and MNE Location Choices

We suggest that the proximity of a country to other countries is a factor that affects its choice as a multinational enterprise (MNE) location. We introduce the concept of a countrys proximity to the global distribution of knowledge, markets, and resources, and frame this concept as a function of both geographic distance and the worldwide spatial distribution of these factors. We test our location model on a data set comprising 138,050 investments undertaken by U.S. MNEs worldwide. Our findings show that the proximity of a country to the rest of the world has a positive impact on MNEs choosing that country as a location. Proximity to the worlds knowledge and markets are stronger drivers of location choice than is proximity to the worlds resources, after accounting for the countrys own endowments. Larger firms are able to benefit more from remote locations than smaller firms are.

Nonparametric Estimation and Regression Analysis with Left-Truncated and Right-Censored Data

Abstract In many prospective and retrospective studies, survival data are subject to left truncation in addition to the usual right censoring. For left-truncated data without covariates, only the conditional distribution of the survival time Y given Y ≥ τ can be estimated nonparametrically, where τ is the lower boundary of the support of the left-truncation variable T. If the data are also right censored, then the conditional distribution can be consistently estimated only at points not larger than τ*, where τ* is the upper boundary of the support of the right-censoring variable C. In this article we first consider nonparametric estimation of trimmed functionals of the conditional distribution of Y, with the trimming inside the observable range between τ and τ*. We then extend the approach to regression analysis and curve fitting in the presence of left truncation and right censoring on the response variable Y. Asymptotic normality of M estimators of the regression parameters derived from this approach is...

Evaluating probability forecasts

Probability forecasts of events are routinely used in climate predictions, in forecasting default probabilities on bank loans or in estimating the probability of a patients positive response to treatment. Scoring rules have long been used to assess the efficacy of the forecast probabilities after observing the occurrence, or nonoccurrence, of the predicted events. We develop herein a statistical theory for scoring rules and propose an alternative approach to the evaluation of probability forecasts. This approach uses loss functions relating the predicted to the actual probabilities of the events and applies martingale theory to exploit the temporal structure between the forecast and the subsequent occurrence or nonoccurrence of the event.

On Asymptotic Power and Efficiency of Tests of Independence in Contingency Tables with Ordered Classifications

Abstract Asymptotic power formulas for tests of independence against simple association models in I × J contingency tables with ordered categories are presented. Yatess score correlation test and Spearmans tied rank correlation test are compared with a likelihood ratio test proposed by S. J. Haberman for testing independence in log-linear models with ordered classifications. Asymptotic power and relative efficiencies are derived under local alternatives that need not be included in the likelihood model assumed by Habermans procedure. Some computations for real and fictitious data are presented.

Hierarchical Dependency Models for Multivariate Survival Data with Censoring

A familyof partial likelihood logistic models is proposed for clusteredsurvival data that are reported in discrete time and that maybe censored. The possible dependence of individual survival timeswithin clusters is modeled, while distinct clusters are assumedto be independent. Two types of clusters are considered. First,all clusters have the same size and are identically distributed.Second, the clusters may vary in size. In both cases our asymptoticresults apply to a large number of small independent clusters.

Controlled ovarian hyperstimulation and transvaginal intratubal insemination as an alternative to gamete intrafallopian transfer

OBJECTIVE To evaluate the efficacy of controlled ovarian hyperstimulation (COH) followed by intratubal insemination in the treatment of infertility. DESIGN Retrospective analysis of 179 intratubal insemination trials in 78 women over a 48-month period. SETTING Reproductive endocrinology practice. PATIENTS Seventy-eight women, 26 to 44 years old (34 +/- 4.3 years; mean +/- SD), classified into subgroups according to diagnosis and age (< 40 or > or = 40 years). INTERVENTIONS Patients underwent COH and intratubal insemination. MAIN OUTCOME MEASURES The rates of pregnancy, delivery, spontaneous abortion, ectopic and multiple gestation, and complications were studied and analyzed statistically. RESULTS Overall, there were 36 pregnancies (20.1% of 179 trials), 29 deliveries (16.2%), 5 of 36 first trimester abortions (13.9%), 2 of 36 ectopic pregnancies (5.6%), and 4 twin gestations. There were no serious complications. In 43 women < 40 years without male factor the first trial resulted in 15 pregnancies (34.9%); in 27 women < 40 years with male factor the first trial resulted in 4 pregnancies (14.8%). Male factor reduced significantly the probability of conception. In 19 women < 40 years of age with either ovarian dysfunction or infertility of unknown cause, the first trial resulted in 9 pregnancies (47.4%), which was significantly better than the 10 pregnancies achieved in the first trial in the remaining 51 women < 40 years of age. Peak serum E2 levels and number of mature follicles had a significant effect on pregnancy rates. CONCLUSIONS Intratubal insemination yielded pregnancy rates comparable to those published for GIFT at reduced cost and with fewer complications.

Confidence limits for small proportions in complex samples

Exact upper confidence limits for small proportions in stratified samples are derived. An algorithm for their computation which employs a new normal approximation for the case of infinitely large strata and a finite number of defectives is proposed. Using selected examples it is shown that the usual confidence limits derived from the standard normal approximation can be highly misleading, and that the exact limits are not unacceptably conservative when compared to natural Empirical Bayes and appropriately defined pseudo-Bayes limits. The loss of efficiency of non-proportionate staratified designs, vis-a-vis simple random sampling or proportionate designs for setting confidence limits on small proportions is studied in a variety of examples. Exact upper confidence limits for small proportions are also derived for simple random samples of equal-size clusters, and a similar algorithm for their derivation is presented. These limits are compared to several Bayes credibility limits in selected examples. The los...

Digital Inequality on Global Online Knowledge Exchange Platforms

The present study reports on an empirical investigation about the major causes of digital inequality in the specific context of user participation on cross-cultural, global knowledge exchange platforms. Using social inclusion theory as a theoretical lens, we propose a predictive research model about digital inequality at the country level and posit that four critical resources, physical resources, digital resources, human resources, and social resources contribute to the effective usage of ICT in accessing, adapting, and creating knowledge. We test our research model on data from a sample of 200 leading knowledge-based platforms.

Multivariate Survival Data With Censoring

We define a new class of models for multivariate survival data, in continuous time, based on a number of cumulative hazard functions, along the lines of our family of models for correlated survival data in discrete time [Gross and Huber-Carol (2000, 2002)]. This family is an alternative to frailty and copula models. We establish some properties of our family and compare it to Clayton’s and Marshall–Olkin’s and derive nonparametric partial likelihood estimates of the hazards involved in its definition.

Bayesian credibility limits for small finite population proportions:stratified and cluster sampling

Bayes credibility limits for small proportions from stratified and fixed size cluster samples are discussed. Ericson’s (JRSS B (1969)) Beta Binomial and Dirichlet-Multinomial priors are used. Approximate limits that are appropriate for large samples and small proportions are derived in both cases. These allow asymptotic comparisons of the efficacy of stratified and cluster sampling relative to simple random sampling for estimating small proportions. Procedures for the selection of hyper parameters are also presented.