Suchergebnisse

The article of record as published may be found at https://digital-commons.usnwc.edu/nwc-review/vol73/iss1/8 ; The predictive power of experts, operations analysis, and the value of information are interwoven subjects that are hard to winnow down to an essence. Prediction is a big subject, so I have limited this article to what I know best: the operational and tactical domains of conventional warfare. First you will read three examples of limitations of predictions when they are formed on the basis of information alone. Next I will demonstrate that even a modest amount of quantitative analysis, even with incomplete information, can help a decision maker execute a military campaign without making explicit predictions about the coming battles or operations. Analysts cannot eliminate wartime surprises, but they can help to avoid the worst mistakes and steer military leaders toward better decisions. I will conclude by advocating what is too rarely done: the comparison of quantitative campaign analysis done before a war with what actually transpired in the war, to show that useful—even critically important—advice can be formulated very quickly to help decision makers. On one hand, intense thinking about the war is necessary; on the other hand, expert judgment alone should be augmented with simple, transparent, timely—even if incomplete—quantitative analysis.

Most empirical policy work focuses on causal inference. We argue an important class of policy problems does not require causal inference but instead requires predictive inference. Solving these "prediction policy problems" requires more than simple regression techniques, since these are tuned to generating unbiased estimates of coefficients rather than minimizing prediction error. We argue that new developments in the field of "machine learning" are particularly useful for addressing these prediction problems. We use an example from health policy to illustrate the large potential social welfare gains from improved prediction.

AbstractWe present a statistical framework for generating predicted dynamic networks based on the observed evolution of social relationships in a population. The framework includes a novel and flexible procedure to sample dynamic networks given a probability distribution on evolving network properties; it permits the use of a broad class of approaches to model trends, seasonal variability, uncertainty, and changes in population composition. Current methods do not account for the variability in the observed historical networks when predicting the network structure; the proposed method provides a principled approach to incorporate uncertainty in prediction. This advance aids in the designing of network-based interventions, as development of such interventions often requires prediction of the network structure in the presence and absence of the intervention. Two simulation studies are conducted to demonstrate the usefulness of generating predicted networks when designing network-based interventions. The framework is also illustrated by investigating results of potential interventions on bill passage rates using a dynamic network that represents the sponsor/co-sponsor relationships among senators derived from bills introduced in the U.S. Senate from 2003 to 2016.

When an employer is found to have made an unlawful threat during an organizing campaign, the National Labor Relations Board typically orders the employer to publicly disavow the threat, which tends to diminish the employer's credibility in the eyes of employees and give the union a leg up in the campaign. In other cases, an employer's illegal threat before a vote on union representation can invalidate the election if the union is defeated at the ballot box.