13 March 2020 in IEEE Access
The mosquito-borne dengue fever is a major public health problem in tropical countries, where it is strongly conditioned by climate factors such as temperature. In this paper, we formulate a holistic machine learning strategy to analyze the temporal dynamics of temperature and dengue data and use this knowledge to produce accurate predictions of dengue, based on temperature on an annual scale. The temporal dynamics are extracted from historical data by utilizing a novel multi-stage combination of auto-encoding, window-based data representation and trend-based temporal clustering. The prediction is performed with a trend association-based nearest neighbour predictor. The effectiveness of the proposed strategy is evaluated in a case study that comprises the number of dengue and dengue hemorrhagic fever cases collected over the period 1985–2010 in 32 federal states of Mexico. The empirical study proves the viability of the proposed strategy and confirms that it outperforms various state-of-the-art competitor methods formulated both in regression and in time series forecasting analysis.
18 November 2018 in Journal of the Royal Statistical Society Series A: Statistics in Society
11 July 2017 in Environmetrics
6 July 2021 in Interest Groups & Advocacy
1 October 2024 in PNAS Nexus
Terrorists and other transnational extremist groups are responsible for thousands of civilian deaths. In confronting extremists, governments have relied heavily on threats, demands, denunciations, and other forms of verbal conflict. Do these efforts at verbal coercion have any effect on terrorist behavior? This analysis focuses on the Islamic State in Iraq and Syria (ISIS), which continues to be the world's deadliest terrorist group and was responsible for recent high-profile attacks in Baghdad, Vienna, Kabul, and Russia. We use Bayesian structural vector autoregression models to analyze daily event data on interactions between ISIS and foreign governments for the 2014–2020 period. We find that verbal conflict initiated by governments not only failed to deter ISIS but in fact increased the frequency of ISIS's attacks on civilians. Additional empirical analyses, combined with evidence from ISIS's publications and public statements, suggest that this effect resulted from a perceived credibility deficit. Extremists use terror attacks to signal that they have the capabilities and willingness to inflict pain and suffering on civilian targets. Government attempts to coerce extremist groups verbally, rather than militarily, reflect an underestimation of the group's capabilities and resolve. In an effort to solidify their reputations, extremists engage in further violence toward civilians, thus leading to worse humanitarian consequences. We extend the analysis to Al-Qaeda in Iraq and Boko Haram and find similar results.
22 May 2019 in The Social Science Journal
The sequencing of the relationship between campaign contributions and legislative rhetoric is categorized by competing theories as being driven by lags, leads, and contemporaneous effects. We study the committee level dynamics between energy sector donations and legislative rhetoric in the U.S. Senate. Conditioning on party and Senate class, we find evidence for complex interactions that are characterized by a combination of various temporal responses. Individual dynamics — electoral vulnerability, geographical differences and the state-specific importance of the sector — lead to stronger connections between campaign contributions and rhetoric.