Alberto Caimo

Knowledge Sharing in Organizations A Bayesian Analysis of the Role of Reciprocity and Formal Structure

We examine the conditions under which knowledge embedded in advice relations is likely to reach across intraorganizational boundaries and be shared between distant organizational members. We emphasize boundary-crossing relations because activities of knowledge transfer and sharing across subunit boundaries are systematically related to desirable organizational outcomes. Our main objective is to understand how organizational and social processes interact to sustain the transfer of knowledge carried by advice relations. Using original fieldwork and data that we have collected on members of the top management team in a multiunit industrial group, we show that knowledge embedded in task advice relations is unlikely to crosscut intraorganizational boundaries, unless advice relations are reciprocated, and supported by the presence of hierarchical relations linking managers in different subunits. The results we report are based on a novel Bayesian Exponential Random Graph Models (BERGMs) framework that allows us to test and assess the empirical value of our hypotheses while at the same time accounting for structural characteristics of the intraorganizational network of advice relations. We rely on computational and simulation methods to establish the consistency of the network implied by the model we propose with the structure of the intraorganizational network that we actually observed.

Efficient computational strategies for doubly intractable problems with applications to Bayesian social networks

Powerful ideas recently appeared in the literature are adjusted and combined to design improved samplers for doubly intractable target distributions with a focus on Bayesian exponential random graph models. Different forms of adaptive Metropolis–Hastings proposals (vertical, horizontal and rectangular) are tested and merged with the delayed rejection (DR) strategy with the aim of reducing the variance of the resulting Markov chain Monte Carlo estimators for a given computational time. The DR is modified in order to integrate it within the approximate exchange algorithm (AEA) to avoid the computation of intractable normalising constant that appears in exponential random graph models. This gives rise to the AEA + DR: a new methodology to sample doubly intractable distributions that dominates the AEA in the Peskun ordering (Peskun Biometrika 60:607–612, 1973) leading to MCMC estimators with a smaller asymptotic variance. The Bergm package for R (Caimo and Friel J. Stat. Softw. 22:518–532, 2014) has been updated to incorporate the AEA + DR thus including the possibility of adding a higher stage proposals and different forms of adaptation.

Simultaneous modeling of initial conditions and time heterogeneity in dynamic networks: An application to Foreign Direct Investments

In dynamic networks, the presence or absence of ties between nodes are subject both to endogenous network dependencies, as well as dependencies stemming from the spatial embedding of nodes. Current statistical models for change over time are typically defined relative to some initial condition, thus skirting the issue of where the first network came from. Additionally, while these longitudinal network models may explain the dynamics of change in the network over time, they do not explain the change in those dynamics. This may be problematic when data are characterized by trends, cycles, and other time-dependent patterns of change. We propose an extension to the longitudinal exponential random graph model that allows for simultaneous inference of the changes over time and the initial conditions, as well as relaxing assumptions of time-homogeneity. Estimation draws on recent Bayesian approaches for cross-sectional exponential random graph models and Bayesian hierarchical models. We develop the model in the context of data on foreign direct investment relations in the global electricity industry during the period 1995-2003. This is a suitable empirical context because international investment relations are known to be affected by factors related to: (i) the initial conditions determined by the geographical location of the countries involved; (ii) timedependent fluctuations in the global intensity of investment flows, and (iii) endogenous network dependencies. We rely on the well-known gravity model used in research on international trade to represent how spatial embedding and endogenous network dependencies jointly shape the dynamics of investment relations.

Bayesian exponential random graph models with nodal random effects

Abstract We extend the well-known and widely used exponential random graph model (ERGM) by including nodal random effects to compensate for heterogeneity in the nodes of a network. The Bayesian framework for ERGMs proposed by Caimo and Friel (2011) yields the basis of our modelling algorithm. A central question in network models is the question of model selection and following the Bayesian paradigm we focus on estimating Bayes factors. To do so we develop an approximate but feasible calculation of the Bayes factor which allows one to pursue model selection. Three data examples and a small simulation study illustrate our mixed model approach and the corresponding model selection.

Management and Follow-up of Patients with a Bronchial Neuroendocrine Tumor in the Last Twenty Years in Ireland: Expected Inconsistencies and Unexpected Discoveries

Bronchial neuroendocrine tumors (NET) are classified into well-differentiated typical carcinoids (TC), atypical carcinoids (AC), large cell neuroendocrine carcinomas (LCNEC), and small cell lung carcinomas (SCLC). We retrospectively reviewed and analyzed the diagnostic and therapeutic aspects, follow-up data, and outcomes of all patients diagnosed with a bronchial NET from 1995 to 2015 at our institution. Patients with LCNEC or SCLC were excluded due to the biological and clinical differences from the other bronchial NET. The clinical, laboratory, imaging, treatment, and follow-up data were collected and analyzed keeping in mind the recently published international recommendations. Forty-six patients were included in the study. Of these, 37 had a TC and 5 an AC. In 4 patients, the histological characterization was inadequate. Forty-four patients underwent surgery. Four patients developed metastatic disease. Interestingly, 14 patients had one or more other tumors diagnosed at some stage and 3 of them had three different tumors. A total of 7 patients died. The analysis of the laboratory and pathology assessment identified some inconsistencies when compared to the international recommendations. Although the treatment of bronchial NET at our institution was consistent with the successively published recommendations, it appears that the diagnostic process and the follow-up surveillance were not. We think that a systematic multidisciplinary approach might improve bronchial NET patient care. A relatively high rate of occurrence of a second, or also a third, non-NET tumor was observed, though the statistical value of such observation could not be exhaustively elucidated in this numerically limited patient population. In our opinion, the observed high rate of second malignancies in this patient cohort highlights the necessity of optimizing the follow-up of the bronchial NET patients, also considering the very good survival rate achieved with regard to the bronchial NET.

Endoscopic Ultrasound Features of Multiple Endocrine Neoplasia Type 1-Related Versus Sporadic Pancreatic Neuroendocrine Tumors: A Single-Center Retrospective Study

Aim: Pancreatic neuroendocrine tumors (pNETs) can occur in patients with a familial syndrome either as multiple endocrine neoplasia type 1 (MEN-1) or as sporadic tumors. Endoscopic ultrasound (EUS) has become one of the first-line investigations for pNET characterization. The ultrasonographic features of pNETs may differ depending on the familial versus sporadic pathogenesis of the tumor. Therefore, the EUS findings could help and direct the definition of a pNET with an impact on the most appropriate diagnostic and therapeutic patient management. Methods: In this single-center retrospective study, we reviewed the EUS features of 94 pNETs from 37 MEN-1 patients and 15 pNETs from 11 sporadic disease patients at the time of their first EUS assessment. We analyzed the most relevant morphological and ultrasonographic characteristics of the tumors and compared the findings between the 2 patient groups. Results: Patients with MEN-1 more likely present with multiple pNETs than patients with sporadic disease. Sporadic pNETs are usually much bigger than those due to MEN-1. Moreover, pNETs are more heterogeneous in patients with sporadic disease than in those with MEN-1. No statistical difference with regard to definition of the margins, morphology, and vascularization of the pNETs appears between the 2 groups. Conclusions: Patients with sporadic disease usually present with bigger and more heterogeneous pNETs than patients with MEN-1, who tend to present with a higher number of lesions. EUS can facilitate the precise characterization of a pNET, and the ultrasonographic features of the lesion can help and distinguish MEN-1-related versus sporadic disease.

Bayesian exponential random graph modelling of interhospital patient referral networks

Using original data that we have collected on referral relations between 110 hospitals serving a large regional community, we show how recently derived Bayesian exponential random graph models may be adopted to illuminate core empirical issues in research on relational coordination among healthcare organisations. We show how a rigorous Bayesian computation approach supports a fully probabilistic analytical framework that alleviates well-known problems in the estimation of model parameters of exponential random graph models. We also show how the main structural features of interhospital patient referral networks that prior studies have described can be reproduced with accuracy by specifying the system of local dependencies that produce - but at the same time are induced by - decentralised collaborative arrangements between hospitals. Copyright