Discrete vs continuous time for large extremes of Gaussian processes.
The academic publication process consists of al least two stages. The first stage covers the conception of a paper, its submission to a journal, possible revisions due to comments made by (anonymous) reviewers, and acceptance of the manuscript, among other aspects. The second stage concerns the eventual publication of the paper and its academic life-cycle, which is usually measured by a citation score. Next to describing this process in some detail, this paper describes the results of an empirical analysis of a database which includes data on a range of aspect of the publication process. Descriptive statistics give insights as to how long it takes (on average) before the editor returns to the author with the reviews, and also how long it takes for the editor to make a final decision on acceptance, based on a revised manuscript. Econometric models are used to see if, for example, the number of pages, the number of pages, the number of authors, and the number of previous rejections have an impact on these times. Also, it is examined if a special issue makes a difference. Finally, it is studied if the editorial process and observable properties of the paper have any effect on the number of citations, which can be seen as a measure of quality.
|Extremes, Gaussian processes, Tail dependence|
|Econometric Institute Research Papers|
|Organisation||Erasmus School of Economics|
Piterbarg, V.I. (2002). Discrete vs continuous time for large extremes of Gaussian processes. (No. EI 2002-06). Econometric Institute Research Papers. Retrieved from http://hdl.handle.net/1765/581