Predicting Job Performance in Small Samples using the Generalized Maximum Entropy Formulation

2012 ◽  
Author(s):  
Pedro J. Ramos-Villagrasa ◽  
Blanca Moreno ◽  
Antonio L. Garcie-Izquierdo
Keyword(s):  
Job Performance ◽  
Maximum Entropy ◽  
Small Samples ◽  
Generalized Maximum Entropy ◽  
Entropy Formulation

scholarly journals Maximum Entropy Analysis of Flow Networks: Theoretical Foundation and Applications

Entropy ◽  
2019 ◽  
Vol 21 (8) ◽  
pp. 776 ◽  
Author(s):  
Robert K. Niven ◽  
Markus Abel ◽  
Michael Schlegel ◽  
Steven H. Waldrip
Keyword(s):  
Maximum Entropy ◽  
Joint Probability ◽  
Financial Economic ◽  
Flow Networks ◽  
Flow Network ◽  
Physical Constraints ◽  
Generalized Maximum Entropy ◽  
Deterministic Solution ◽  
Network Properties ◽  
The Cost

The concept of a “flow network”—a set of nodes and links which carries one or more flows—unites many different disciplines, including pipe flow, fluid flow, electrical, chemical reaction, ecological, epidemiological, neurological, communications, transportation, financial, economic and human social networks. This Feature Paper presents a generalized maximum entropy framework to infer the state of a flow network, including its flow rates and other properties, in probabilistic form. In this method, the network uncertainty is represented by a joint probability function over its unknowns, subject to all that is known. This gives a relative entropy function which is maximized, subject to the constraints, to determine the most probable or most representative state of the network. The constraints can include “observable” constraints on various parameters, “physical” constraints such as conservation laws and frictional properties, and “graphical” constraints arising from uncertainty in the network structure itself. Since the method is probabilistic, it enables the prediction of network properties when there is insufficient information to obtain a deterministic solution. The derived framework can incorporate nonlinear constraints or nonlinear interdependencies between variables, at the cost of requiring numerical solution. The theoretical foundations of the method are first presented, followed by its application to a variety of flow networks.

scholarly journals A Maximum-Entropy Method to Estimate Discrete Distributions from Samples Ensuring Nonzero Probabilities

Entropy ◽  
2018 ◽  
Vol 20 (8) ◽  
pp. 601 ◽  
Author(s):  
Paul Darscheid ◽  
Anneli Guthke ◽  
Uwe Ehret
Keyword(s):  
Maximum Entropy ◽  
Multinomial Distribution ◽  
Entropy Method ◽  
Small Sample ◽  
Discrete Distributions ◽  
Occupation Probability ◽  
Small Samples ◽  
Data Set ◽  
Sample Distribution ◽  
Leibler Divergence

When constructing discrete (binned) distributions from samples of a data set, applications exist where it is desirable to assure that all bins of the sample distribution have nonzero probability. For example, if the sample distribution is part of a predictive model for which we require returning a response for the entire codomain, or if we use Kullback–Leibler divergence to measure the (dis-)agreement of the sample distribution and the original distribution of the variable, which, in the described case, is inconveniently infinite. Several sample-based distribution estimators exist which assure nonzero bin probability, such as adding one counter to each zero-probability bin of the sample histogram, adding a small probability to the sample pdf, smoothing methods such as Kernel-density smoothing, or Bayesian approaches based on the Dirichlet and Multinomial distribution. Here, we suggest and test an approach based on the Clopper–Pearson method, which makes use of the binominal distribution. Based on the sample distribution, confidence intervals for bin-occupation probability are calculated. The mean of each confidence interval is a strictly positive estimator of the true bin-occupation probability and is convergent with increasing sample size. For small samples, it converges towards a uniform distribution, i.e., the method effectively applies a maximum entropy approach. We apply this nonzero method and four alternative sample-based distribution estimators to a range of typical distributions (uniform, Dirac, normal, multimodal, and irregular) and measure the effect with Kullback–Leibler divergence. While the performance of each method strongly depends on the distribution type it is applied to, on average, and especially for small sample sizes, the nonzero, the simple “add one counter”, and the Bayesian Dirichlet-multinomial model show very similar behavior and perform best. We conclude that, when estimating distributions without an a priori idea of their shape, applying one of these methods is favorable.

scholarly journals Effect of Actual and Perceived Violence on Internal Migration: Evidence from Mexico’s Drug War

2020 ◽  
Vol 11 (1) ◽  
Author(s):  
Amilcar Orlian Fernandez-Dominguez
Keyword(s):  
Maximum Entropy ◽  
Gravity Model ◽  
Internal Migration ◽  
Interstate Migration ◽  
Limited Evidence ◽  
Drug War ◽  
Homicide Rates ◽  
Oaxaca Decomposition ◽  
Generalized Maximum Entropy ◽  
Mexican Drug War

AbstractAccording to the Organisation for Economic Co-operation and Development (OECD), violence should be considered by examining both actual and perceived crime. However, the studies related to violence and internal migration under the Mexican drug war episode focus only on one aspect of violence (perception or actual), so their conclusions rely mostly on limited evidence. This article complements previous work by examining the effects of both perceived and actual violence on interstate migration through estimation of a gravity model along three 5-year periods spanning from 2000 to 2015. Using the methods of generalized maximum entropy (to account for endogeneity) and the Blinder–Oaxaca decomposition, the results show that actual violence (measured by homicide rates) does affect migration, but perceived violence explains a greater proportion of higher average migration after 2005. Since this proportion increased after 2010 and actual violence, the results suggest that there was some adaptation to the new levels of violence in the period 2010–2015.

Business Firms' Responses to the Crises of 2009

2014 ◽  
Vol 5 (1) ◽  
pp. 92-110
Author(s):  
Sean M. Murphy ◽  
Daniel L. Friesner ◽  
Robert Rosenman
Keyword(s):  
Maximum Entropy ◽  
Economic Crisis ◽  
Influenza Pandemic ◽  
Economic Uncertainty ◽  
Business Managers ◽  
Generalized Maximum Entropy ◽  
Influenza Outbreaks

In 2009 firms faced both economic uncertainty and influenza outbreaks. Both crises posed large costs for firms; however, the manner in which they were perceived by management to affect the organization potentially differed. Using generalized maximum entropy (GME) the authors analyzed a business outlook survey of Seattle, Washington area businesses. Overall, firms were more proactive in responding to the economic crisis than to the influenza pandemic, even though the potential costs associated with both were quite large. Among the authors' conclusions is that business managers responded to the economic crisis more because it was more familiar and something over which they thought they had more control.

scholarly journals Generalized Versus Classical Maximum Entropy for Imaging

1994 ◽  
Vol 158 ◽  
pp. 215-217
Author(s):  
A.T. Bajkova
Keyword(s):  
Maximum Entropy ◽  
Maximum Entropy Method ◽  
Fourier Spectrum ◽  
Careful Analysis ◽  
Entropy Method ◽  
Reconstruction Method ◽  
Generalized Maximum Entropy ◽  
Exact Reconstruction ◽  
Reconstruction Quality ◽  
Complex Functions

The problem of image reconstruction from incomplete and noisy complex Fourier spectrum is considered. The maximum entropy method (MEM) is of great interest as the most effective nonlinear reconstruction method having superresolution effect. Because objects of radio astronomical observations are incoherent radio sources described by real non-negative distributions, application of the classical MEM is quite reasonable. But it is established that the MEM gives acceptable reconstruction quality mostly in the case of point-like sources and in general it does not ensure satisfactory reconstruction of continous, graytone objects, which can considerably restrict applications of the MEM in astronomy. The generalized maximum entropy method (GMEM) was originally proposed for reconstruction of distributions described by complex functions (Bajkova, 1992) and was considered as having the same properties of the classical MEM. More careful analysis of the GMEM and classical MEM for real non-negative objects allowed to establish that the GMEM ensures much more exact reconstruction, especially in the case of continous objects. Explanation and demonstration of this interesting and very important phenomenon is the purpose of the present paper.

Sign in / Sign up

Export Citation Format

Share Document