Search

IntroductionCOVID-19 has caused tremendous death and suffering since it first emerged in 2019. Soon after its emergence, models were developed to help predict the course of various disease metrics, and these models have been relied upon to help guide public health policy.MethodsHere we present a method called COVIDNearTerm to "forecast" hospitalizations in the short term, two to four weeks from the time of prediction. COVIDNearTerm is based on an autoregressive model and utilizes a parametric bootstrap approach to make predictions. It is easy to use as it requires only previous hospitalization data, and there is an open-source R package that implements the algorithm. We evaluated COVIDNearTerm on San Francisco Bay Area hospitalizations and compared it to models from the California COVID Assessment Tool (CalCAT).ResultsWe found that COVIDNearTerm predictions were more accurate than the CalCAT ensemble predictions for all comparisons and any CalCAT component for a majority of comparisons. For instance, at the county level our 14-day hospitalization median absolute percentage errors ranged from 16 to 36%. For those same comparisons, the CalCAT ensemble errors were between 30 and 59%.ConclusionCOVIDNearTerm is a simple and useful tool for predicting near-term COVID-19 hospitalizations.

The feasibility of the single-phase, full-bridge rectifier with an LC filter to meet the IEC 1000-3-2, class-D specifications is assessed. It is found that this passive LC-filter approach can meet the required specifications if a proper inductance value of the filter choke is selected. Choke design considerations and performance evaluation results that include the power loss, volume, and weight estimates for applications with power levels between 75 and 600 W are presented. Index Terms - Electromagnetic compatibility, IEC 1000-3-2 specifications, passive input-current shaping, passive power-factor correction, rectifier with LC filter.

A simple, low-cost, and robust 'democratic' (autonomous) current-sharing (CS) circuit is proposed and analyzed. The circuit maintains good CS among the modules by properly adjusting the voltage references of the modules based on the differences between the average current of the paralleled modules and the currents of the individual modules. Design guidelines for achieving a desired CS performance are given and verified on a number of dc/dc modules operating in parallel.

Transparency guarantees in Mexico presented a serious challenge to its 2007–2012 war on drugs. We use an original database of access to information requests, including both petitions for information and appeals to IFAI. We conduct statistical tests on the databases, finding that transparency is lower on security issues, as expected, but that there are unexpected variations between security agencies, and over time. We then conduct a content analysis of freedom of information requests to determine what drives agency responses, finding that security agencies developed various techniques to deflect petitions for information, such as falsely claiming that information has been provided when it has not, claiming that the information is outside the competence of the agency, that it does not exist, or that it is already in the public domain. Also there are significant differences in transparency between security agencies, possibly explained by their operational roles. [ABSTRACT FROM AUTHOR]

Mulling over politics with others can change citizens’ political beliefs and choices. Is the effect of interpersonal political discussion different when one of the interlocutors has a family member living abroad—that is, is a “transnational household member” (THM)? Using data from 20 Latin American countries in the 2006-2008 AmericasBarometer, we show that talking about politics makes THMs less satisfied with their democracies and less proud of their political systems than non-THMs. When THMs engage in cross-border political discussions with relatives abroad, they gain new information and perspectives that cast their own democracy in a different light. Even absent cross-border communication, though, political discussion with peers at home can make THMs more critical by emphasizing their government’s transnational governance obligations—and highlighting failures to live up to these heightened expectations. Our study thus adds to a growing body of research on emigration’s impact on sending country politics. [ABSTRACT FROM AUTHOR]

Just as Mexico—particularly civil society—has joined the rights revolution, a crime wave is convulsing the country. Local human rights organizations (LHROs) are crucial intercessors on behalf of citizens. Public trust is vital for LHROs to be effective, especially in their tasks of promoting police and justice reform. How does crime affect trust in LHROs? Does it reduce trust, leading to a “rights trap” in which lack of citizen support wrests legitimacy from LHROs and governments retrench on rights? Or could itincreaseLHRO trust, leading to a “rights amplifier” in which greater LHRO efficacy fosters rights observance? Using data from the 2014The Americas and the World/Human Rights Perceptions Polls, I find that crime's effects vary according to the security and institutional context. Victimization damages LHRO trust in high-crime communities but enhances it in low-crime ones. Crime can trigger both rights traps and amplifiers. [ABSTRACT FROM PUBLISHER]