Partial least squares Cox regression for genome-wide data

Most methods for survival prediction from high-dimensional genomic data combine the Cox proportional hazards model with some technique of dimension reduction, such as partial least squares regression (PLS). Applying PLS to the Cox model is not entirely straightforward, and multiple approaches have been proposed. The method of Park et al. (Bioinformatics 18(Suppl. 1):S120–S127, 2002) uses a reformulation of the Cox likelihood to a Poisson type likelihood, thereby enabling estimation by iteratively reweighted partial least squares for generalized linear models. We propose a modification of the method of park et al. (2002) such that estimates of the baseline hazard and the gene effects are obtained in separate steps. The resulting method has several advantages over the method of park et al. (2002) and other existing Cox PLS approaches, as it allows for estimation of survival probabilities for new patients, enables a less memory-demanding estimation procedure, and allows for incorporation of lower-dimensional non-genomic variables like disease grade and tumor thickness. We also propose to combine our Cox PLS method with an initial gene selection step in which genes are ordered by their Cox score and only the highest-ranking k% of the genes are retained, obtaining a so-called supervised partial least squares regression method. In simulations, both the unsupervised and the supervised version outperform other Cox PLS methods.     

Discounting with fat-tailed economic growth

When the growth of aggregate consumption exhibits no serial correlation, the socially efficient discount rate is independent of the time horizon, because the wealth effect and the precautionary effect are proportional to the time horizon. In this paper, we consider alternative growth processes: an AR(1), a Brownian motion with unknown trend or volatility, a two-state regime-switching model, and a model with an uncertain return of capital. All these models exhibit some persistence of shocks on the growth rate of the economy and fat tails, which implies that one should discount more distant costs and benefits at a smaller rate.     

Supply Management Under High Goal Incongruence: An Empirical Examination of Disintermediation in the Aerospace Supply Chain

Aftermarket sales and profits are becoming an increasingly important part of an original equipment manufacturer's (OEM) business model. Because replacement parts often do not require further manufacturing, OEMs act as intermediaries in the aftermarket. As with any intermediary, the OEM must concern itself with suppliers disintermediating its supply chain selling replacement parts directly to the OEM's customers. We frame supply chain disintermediation (SCD) as a principal–agent contracting problem between an OEM buyer and a supplier. Hypotheses relate contract conditions, goal incongruence, supplier capabilities and contract enforcement to SCD. The data are collected from the aerospace industry using a multimethod study, combining an Internet‐based survey with archival data. Causal modeling with structural equation modeling (SEM) shows general support for the hypotheses. Particularly, SCD is positively related to buyer–supplier goal incongruence. The agency model offers insights that differ from previous transaction‐cost‐based models of buyer–supplier relationships. OEM buyers with a lucrative aftermarket should consider aligning goals through incentives rather than relying entirely on economic hostages associated with specific assets.     

Erfassung betriebswirtschaftlich relevanter Zeitschriften in den ISI-Datenbanken sowie der Scopus-Datenbank

Zusammenfassung  Publikations- und Zitationsindizes k?nnen insbesondere zur Quellenerschlie?ung und Forschungsevaluation genutzt werden. Die Eignung entsprechender Indizes zur ad?quaten Erfüllung dieser Funktionen h?ngt ma?geblich von der Auswahl der indexierten Publikationen ab. In diesem Beitrag wird die Publikationsauswahl des Social Sciences Citation Index (SSCI), des Science Citation Index-Expanded (SCI-X) und der Datenbank Scopus aus Sicht der deutschsprachigen Betriebswirtschaftslehre analysiert. Als Basis für die Untersuchung dienen die Ergebnisse des VHB-JOURQUAL. Neben einem Gesamtüberblick zur Erfassung betriebswirtschaftlich relevanter Zeitschriften in den Datenbanken wird auch die Abdeckung einzelner betriebswirtschaftlicher Forschungsfelder untersucht. Da die Nutzung der Indizes mit einigen Kosten für den Erwerb entsprechender Lizenzen verbunden ist, erfolgt darüber hinaus eine vergleichende Analyse von SSCI, SCI-X und Scopus hinsichtlich der Zeitschriftenabdeckung. Aufgrund der internationalen Ausrichtung der betrachteten Zitationsindizes wird die Untersuchung zus?tzlich auf ein internationales Zeitschriftenranking ausgedehnt.

---

Acquisition of economic journals in the ISI-databases and in Scopus

Summary  Publication- and citation-indices are especially used for searching literature and evaluating research performance. Their suitability strongly depends on the publications taken into account. In this paper we analyse the selection of publications in Social Sciences Citation Index (SSCI), Science Citation Index-Expanded (SCI-X) and Scopus for G erman speaking researchers of business economics. Therewith, the VHB-JOURQUAL provides a basis for our analysis. In addition to an overview of economic relevant journals in the indices we analyse the coverage of special fields of business economics. Due to the fact that using such indices is very expensive, we compare SSCI, SCI-X and Scopus concerning their coverage of journals. As the indices are of international orientation, we enlarge the analysis onto an international ranking of publications.

     

Alignments with non-overlapping moves,inversions and tandem duplications in <Emphasis Type="Italic">O</Emphasis>(<Emphasis Type="Italic">n</Emphasis><Superscript>4</Superscript>) time

Sequence alignment is a central problem in bioinformatics. The classical dynamic programming algorithm aligns two sequences by optimizing over possible insertions, deletions and substitutions. However, other evolutionary events can be observed, such as inversions, tandem duplications or moves (transpositions). It has been established that the extension of the problem to move operations is NP-complete. Previous work has shown that an extension restricted to non-overlapping inversions can be solved in O(n ³) with a restricted scoring scheme. In this paper, we show that the alignment problem extended to non-overlapping moves can be solved in O(n ⁵) for general scoring schemes, O(n ⁴log n) for concave scoring schemes and O(n ⁴) for restricted scoring schemes. Furthermore, we show that the alignment problem extended to non-overlapping moves, inversions and tandem duplications can be solved with the same time complexities. Finally, an example of an alignment with non-overlapping moves is provided. A preliminary version of this paper appeared in the Proceedings of COCOON 2007, LNCS, vol. 4598, pp. 151–164.     

Finding socially best spanning trees

This article combines Social Choice Theory with Discrete Optimization. We assume that individuals have preferences over edges of a graph that need to be aggregated. The goal is to find a socially “best” spanning tree in the graph. As ranking all spanning trees is becoming infeasible even for small numbers of vertices and/or edges of a graph, our interest lies in finding algorithms that determine a socially “best” spanning tree in a simple manner. This problem is closely related to the minimum (or maximum) spanning tree problem in Discrete Optimization. Our main result shows that for the various underlying ranking rules on the set of spanning trees discussed in this article, the sets of “best” spanning trees coincide. Moreover, a greedy algorithm based on a transitive group ranking on the set of edges will always provide such a “best” spanning tree.     

The Termination of Public Organizations: Theoretical Perspectives to Revitalize a Promising Research Area

While many studies deal with comparative public sector reform, the fundamental question of whether and to what extent states are actually able to abolish parts of their administrative structure remains untackled. Despite some efforts to solve this puzzle, the topic remains underestimated. This article identifies the main conceptual and theoretical problems associated with existent research on the termination of public organizations. Furthermore, the article systemizes various causal factors of termination into two broad dimensions: “organizational stickiness” and “political incentives.” Taken together, these constitute a typology, which is able to guide future empirical investigation of the termination of public organizations.