Longitudinal Data Analysis and Time Series Commons^™

Stock Forecasts With Lstm And Web Sentiment, Michael Burgess, Faizan Javed, Nnenna Okpara, Chance Robinson

SMU Data Science Review

Traditional time-series techniques, such as auto-regressive and moving average models, can have difficulties when applied to stock data due to the randomness inherent to the markets. In this study, Long Short-Term Memory Recurrent Neural Networks, or LSTMs, have been applied to pricing data along with sentiment scores derived from web sources such as Twitter and other financial media outlets. The project team utilized this approach to complement the technical indicators observed at the end of each trading day for three stocks from the NASDAQ stock exchange over a 12-year span. A common benchmark to assess model performance on time series …

A Transformer-Based Classification System For Volcanic Seismic Signals, Anthony P. Rinaldi, Cindy Mora Stock, Cristián Bravo Roman, Alexander Hemming

Undergraduate Student Research Internships Conference

Monitoring volcanic events as they occur is a task that, to this day, requires significant human capital. The current process requires geologists to monitor seismographs around the clock, making it extremely labour-intensive and inefficient. The ability to automatically classify volcanic events as they happen in real-time would allow for quicker responses to these events by the surrounding communities. Timely knowledge of the type of event that is occurring can allow these surrounding communities to prepare or evacuate sooner depending on the magnitude of the event. Up until recently, not much research has been conducted regarding the potential for machine learning …

Functional Structure Of Excess Return And Volatility, Chenxi Zhao

Undergraduate Student Research Internships Conference

Capturing the relation between excess returns and volatility can help making better decisions in the stock market in terms of portfolio allocation and assets risk management. This paper takes the data of a minute-by-minute series of S&P500 from January 2009 to January 2021 as the research object and explores the best structural representation for the excess return as a function of the volatility, for a well-known index. This is implemented via regression models for volatility and excess returns. The results reveal that there’s a structural break in the relationship between the excess return and volatility based on the sign of …

Dynamic Prediction For Alternating Recurrent Events Using A Semiparametric Joint Frailty Model, Jaehyeon Yun

Statistical Science Theses and Dissertations

Alternating recurrent events data arise commonly in health research; examples include hospital admissions and discharges of diabetes patients; exacerbations and remissions of chronic bronchitis; and quitting and restarting smoking. Recent work has involved formulating and estimating joint models for the recurrent event times considering non-negligible event durations. However, prediction models for transition between recurrent events are lacking. We consider the development and evaluation of methods for predicting future events within these models. Specifically, we propose a tool for dynamically predicting transition between alternating recurrent events in real time. Under a flexible joint frailty model, we derive the predictive probability of …

Data-Driven Analysis Of Drug And Substance Abuse Rates Across The Varying Regions In The United States Of America, Reem Saleh

University Honors Theses

Drugs and substance abuse is one of the leading causes of death for adolescents in the United States. The consequences of using these drugs are profound and can cause both damage to one's physical and psychological health. The rates of drug abuse in the United States continue to increase over the years. This paper analyzes the trends in rates of drug abuse in the four regions in the United States. It looks at the rates in cocaine, cigarettes, marijuana, and tobacco. A preliminary analysis was done to look at the trend in rates followed by an ARIMA time series model …

A Novel Correction For The Adjusted Box-Pierce Test, Sidy Danioko, Jianwei Zheng, Kyle Anderson, Alexander Barrett, Cyril S. Rakovski

Mathematics, Physics, and Computer Science Faculty Articles and Research

The classical Box-Pierce and Ljung-Box tests for auto-correlation of residuals possess severe deviations from nominal type I error rates. Previous studies have attempted to address this issue by either revising existing tests or designing new techniques. The Adjusted Box-Pierce achieves the best results with respect to attaining type I error rates closer to nominal values. This research paper proposes a further correction to the adjusted Box-Pierce test that possesses near perfect type I error rates. The approach is based on an inflation of the rejection region for all sample sizes and lags calculated via a linear model applied to simulated …

Intervention Time Series Analysis Of Organ Donor Transplants In The Us, Supraja Malladi

Biology and Medicine Through Mathematics Conference

No abstract provided.

Sparse Model Selection Using Information Complexity, Yaojin Sun

Doctoral Dissertations

This dissertation studies and uses the application of information complexity to statistical model selection through three different projects. Specifically, we design statistical models that incorporate sparsity features to make the models more explanatory and computationally efficient.

In the first project, we propose a Sparse Bridge Regression model for variable selection when the number of variables is much greater than the number of observations if model misspecification occurs. The model is demonstrated to have excellent explanatory power in high-dimensional data analysis through numerical simulations and real-world data analysis.

The second project proposes a novel hybrid modeling method that utilizes a mixture …

Penalized Estimation Of Autocorrelation, Xiyan Tan

All Dissertations

This dissertation explored the idea of penalized method in estimating the autocorrelation (ACF) and partial autocorrelation (PACF) in order to solve the problem that the sample (partial) autocorrelation underestimates the magnitude of (partial) autocorrelation in stationary time series. Although finite sample bias corrections can be found under specific assumed models, no general formulae are available. We introduce a novel penalized M-estimator for (partial) autocorrelation, with the penalty pushing the estimator toward a target selected from the data. This both encapsulates and differs from previous attempts at penalized estimation for autocorrelation, which shrink the estimator toward the target value of zero. …

Deep Depression Prediction On Longitudinal Data Via Joint Anomaly Ranking And Classification, Guansong Pang, Ngoc Thien Anh Pham, Emma Baker, Rebecca Bentley, Anton Van Den Hengel

Research Collection School Of Computing and Information Systems

A wide variety of methods have been developed for identifying depression, but they focus primarily on measuring the degree to which individuals are suffering from depression currently. In this work we explore the possibility of predicting future depression using machine learning applied to longitudinal socio-demographic data. In doing so we show that data such as housing status, and the details of the family environment, can provide cues for predicting future psychiatric disorders. To this end, we introduce a novel deep multi-task recurrent neural network to learn time-dependent depression cues. The depression prediction task is jointly optimized with two auxiliary anomaly …

Intra-Hour Solar Forecasting Using Cloud Dynamics Features Extracted From Ground-Based Infrared Sky Images, Guillermo Terrén-Serrano

Electrical and Computer Engineering ETDs

Due to the increasing use of photovoltaic systems, power grids are vulnerable to the projection of shadows from moving clouds. An intra-hour solar forecast provides power grids with the capability of automatically controlling the dispatch of energy, reducing the additional cost for a guaranteed, reliable supply of energy (i.e., energy storage). This dissertation introduces a novel sky imager consisting of a long-wave radiometric infrared camera and a visible light camera with a fisheye lens. The imager is mounted on a solar tracker to maintain the Sun in the center of the images throughout the day, reducing the scattering effect produced …

Are Long-Period Exoplants Around Cool Stars More Common Than We Thought?, Emily Jane Safron

LSU Doctoral Dissertations

The Kepler mission has been the catalyst for discovery of nearly 5,000 confirmed and candidate exoplanets. The majority of these candidates orbit Sun-like stars, and have orbital periods comparable to or shorter than that of the Earth, due to the selection bias inherent in the transit method and the limitations of automated transit search algorithms. We aim to develop a richer understanding of the population of exoplanets around the lowest-mass stars, the M spectral type. We are particularly interested in exoplanets with long orbital periods, which are difficult or impossible to find using standard transit search algorithms. In our study, …

Impact Of Loss To Follow-Up And Time Parameterization In Multiple-Period Cluster Randomized Trials And Assessing The Association Between Institution Affiliation And Journal Publication, Jonathan Moyer

Doctoral Dissertations

Difference-in-difference cluster randomized trials (CRTs) use baseline and post-test measurements. Standard power equations for these trials assume no loss to follow-up. We present a general equation for calculating treatment effect variance in difference-in-difference CRTs, with special cases assuming loss to follow-up with replacement of lost participants and loss to follow-up with no replacement but retaining the baseline measurements of all participants.

Multiple-period CRTs can represent time as continuous using random coefficients (RC) or categorical using repeated measures ANOVA (RM-ANOVA) analytic models. Previous work recommends the use of RC over RM-ANOVA for CRTs with more than two periods because RC exhibited …

Slices Of The Big Apple: A Visual Explanation And Analysis Of The New York City Budget, Joanne Ramadani

Dissertations, Theses, and Capstone Projects

As a component of government, budgets are fundamental not only to improving the quality of a shared society, but also to understanding what our government officials consider to be their priorities. However, most budgets can be difficult to understand, using terms that are not familiar to people who have not studied finance or economics. To that end, Slices of the Big Apple is an interactive, centralized narrative website that uses visualizations at its core in order to: 1) facilitate a holistic understanding of the New York City government budget for NYC residents; and 2) conduct a five-year analysis of Community …

Análisis De Los Días De Mora Para La Cartera De Un Producto Financiero En Colombia, Una Aproximación A Partir De Las Series De Tiempo (2013 - 2018), Eleny Kottaridis Fernandez

Economía

La morosidad sobre la cartera de consumo evidencia un patrón que debe ser considerado en la toma de decisiones de las entidades financieras para una adecuada administración del riesgo crediticio teniendo en cuenta su alta volatilidad. En efecto, un desempeño económico desfavorable relacionado con algunos sectores financieros, las bajas tasas de crecimiento económico y mayores niveles de desempleo, incrementa la probabilidad del incumplimiento de las obligaciones de los hogares debido a la menor capacidad de pago por la reducción de sus ingresos. De acuerdo con estos impactos, las entidades financieras necesitan contar con mecanismos para abordar el pronóstico sobre el …

Estimating The Statistics Of Operational Loss Through The Analyzation Of A Time Series, Maurice L. Brown

Theses and Dissertations

In the world of finance, appropriately understanding risk is key to success or failure because it is a fundamental driver for institutional behavior. Here we focus on risk as it relates to the operations of financial institutions, namely operational risk. Quantifying operational risk begins with data in the form of a time series of realized losses, which can occur for a number of reasons, can vary over different time intervals, and can pose a challenge that is exacerbated by having to account for both frequency and severity of losses. We introduce a stochastic point process model for the frequency distribution …

Lake Huron Shoreline Analysis, Shubham Satish Nandanwar

Theses and Dissertations (Comprehensive)

Lake Huron is a popular tourist destination and is home to several businesses and residents. Since the shoreline is dynamic and is subject to change over the years due to several factors such as a change in water level, soil type, human encroachment, etc., these locations tend to encounter floods due to increased water levels and wind speed. This causes erosion and loss to the properties along the shoreline.

This study is based on two areas of interest named Pinery Provincial Park and Sauble Beach which are located on the shoreline of Lake Huron where Pinery Provincial Park is a …

Predicting Power Using Time Series Analysis Of Power Generation And Consumption In Texas, Joshua Eysenbach, Bodie Franklin, Andrew J. Larsen, Joel Lindsey

SMU Data Science Review

Due to the recent power events in Texas, power forecasting has been brought national attention. Accurate demand forecasting is necessary to be sure that there is adequate power supply to meet consumer's needs. While Texas has a forecasting model created by the Electricity Reliability Council of Texas (ERCOT), constant efforts are required to ensure that the model stays at the state-of-the-art and is producing the most reliable forecasts possible. This research seeks to provide improved short- and medium-term forecasting models, bringing in state-of-the-art deep learning models to compare to ERCOT’s forecasts. A model that is more accurate than ERCOT’s own …

Bayesian Variable Selection Strategies In Longitudinal Mixture Models And Categorical Regression Problems., Md Nazir Uddin

Electronic Theses and Dissertations

In this work, we seek to develop a variable screening and selection method for Bayesian mixture models with longitudinal data. To develop this method, we consider data from the Health and Retirement Survey (HRS) conducted by University of Michigan. Considering yearly out-of-pocket expenditures as the longitudinal response variable, we consider a Bayesian mixture model with $K$ components. The data consist of a large collection of demographic, financial, and health-related baseline characteristics, and we wish to find a subset of these that impact cluster membership. An initial mixture model without any cluster-level predictors is fit to the data through an MCMC …

Multiple Baseline Interrupted Time Series: Describing Changes In New Mexico Medicaid Behavioral Health Home Patients’ Care, Jessica Reno

Mathematics & Statistics ETDs

In 2016, the CareLink New Mexico behavioral health homes program began enrolling Medicaid recipients with the goal of increasing care coordination, improving access to services, and decreasing long-term costs of care for adults with serious mental illness (SMI) and children with severe emotional disturbance (SED). To evaluate these aims, a retrospective interrupted time series study using Medicaid claims data was designed. First, a comparable subset of non-enrolled individuals was selected from the pool of Medicaid recipients with SMI or SED using propensity score matching. Then, segmented regression was applied to three outcomes: total Medicaid charges, number of outpatient behavioral health …