Presentations
2024
- Refereed talk at SDSS 2024 View slides in full screen
2023
- Colloquium Talk at Cal Poly Humboldt Math Department: Online Trend Estimation and Detection of Trend Deviations in Sub-sewershed Time Series of SARS-CoV-2 RNA Measured in Wastewater
Wastewater surveillance has proven a key public health tool to understand a wide range of community health diseases and has been especially critical to health departments throughout the SARS CoV-2 pandemic. Regular sampling of wastewater at routinely monitored locations is cost effective and gives a time series of viral concentration of SARS-CoV-2. When time series data are collected, the goal is often to estimate a trend, that is, whether the “typical values” are changing in time. When visualized, it is clear that wastewater viral concentrations are changing in time and can be modeled using time series tools. Such a model should be able to separate out the “noise”, or observation/measurement error, in these observations from the “signal”, or trend. An additional constraint when modeling time series data is the presence of temporal correlation structure, i.e. the values are not independent. The state space modeling framework is quite flexible and can separate out sources of variation while also accounting for temporal dependence structures. This talk will detail the application of a state space model and Exponentially Weighted Moving Average control charts to several wastewater time series with the goal of determining when the upstream sub-sewersheds are deviating from the routinely monitored, centralized WWTP. The results of these analyses can be used to inform public health resource allocation decisions in times of increasing concern.
- Conference Talk: Spatial-Temporal Extreme Modeling through Point-to-Area Random Effects (PARE)
One measurement modality for rainfall is a fixed location rain gauge. However, it is important to evaluate extreme rainfall, flooding, and other climate extremes at a larger spatial scale. These phenomenon often occur at larger scales and affect more than one location in a community. For example, in 2017 Hurricane Harvey impacted all of Houston and the surrounding region causing widespread flooding. Flood risk modeling requires understanding of rainfall for hydrolo44 gic regions. Further, policy changes to address the risks and damages of natural hazards such as severe flooding are usually made at the community/neighborhood level or higher. Therefore, spatial-temporal methods which convert results from one spatial scale to another are especially useful in applications for evolving environmental extremes. We develop modeling strategy for understanding spatial-temporal extreme values at the areal level, when the core information are time series at point locations distributed over the region.
2022
Lost to corporate archives.
2021
Lost to corporate archives.
2020
- Conference Talk: A spatiotemporal case crossover model of asthma attacks in the City of Houston,SDSS 2020
Case crossover design is a modeling approach used to assess the effect of a transient exposure on an acute outcome. Case crossover has been applied to many public health challenges that often involve spatially referenced data, but do not typically model any kind of spatial dependence due to the case crossover assumption. Based on an existing equivalence between case crossover analyses and time series count regression, this paper applies a case crossover model in time with a spatial random effect based on a Hausdorff-distance based weight matrix which accounts for the geometry of areal units to study asthma attacks in the City of Houston in 2015.
2019
- Thesis Defense: Advances in the Analysis of Spatially Aggregated Data
An understanding of the spatial relationships in sociological and epidemiological applications is an important tool in the analysis of urban data. While point level data (e.g. observations at a given latitude/longitude) provide the most detail about spatial phenomenon, spatial data aggregated to the level of relevant municipal regions is easily accessible and can provide insights at a level useful for policy decisions for governments and communities. This work identifies two areas of focus in the analysis of spatially aggregated data. First, a new specification for dependence in spatial regression models for aggregated data using the Hausdorff distance and extended Hausdorff distance is introduced. The new dependence structure is shown to account for the shape and orientation of the irregular and disconnected regions often encountered in practice and evaluated in the context of model performance as well as a real data example. An R package compatible with existing spatial packages which implements the construction of spatial weight matrices generated using the (extended) Hausdorff distance is provided along with a vignette illustrating its use on real data. Second, the idea of a spatial case-crossover model is explored in the context of connection to existing spatial methods. A method for including spatial dependence in a spatio-temporal case-crossover model is also explored.
- Conference Talk: (given jointly with Katherine Ensor) Understanding Urban Pollution Through Spatial-Temporal Modeling
Spatial-temporal modeling of pollution in an urban environment is key to helping city and community officials manage the impact of pollution. In this talk we will explore the dynamic structure of groundwater pollution after the historical flood of Houston caused by Hurricane Harvey in 2017. The statisical innovation incorporates a distant metric that encompasses the geography of the region. Using Hausdorff distance we establish an exposure index from this historical flood for residents in Houston. The index values will be published through the Kinder Institute Urban Data Platform (kinderudp.org). Further, linking our index to the socio-economic data available on this platform we are able to quantify differential population exposure, pinpointing those residents at highest risk. This information is then used by city and community officials to mitigate the consequences of the aberrant pollution.