DigitalGeorgetown
Not a member yet
425578 research outputs found
Sort by
Essays on Latent Variables in Time-Series and Panel Econometrics
No full textPh.D.In the dissertation, I analyze both empirical and theoretical topics related to the existence of latent variables. In Chapter 1, I propose a non-linear time-series model incorporating a particular neural network algorithm, a recurrent neural network. The construction of neural network algorithms shows the models approximate the data process by taking advantage of latent hidden units. A proposed model combines an economic model-based restriction with the recurrent neural network algorithm, where serial correlation is considered and the results are interpretable in terms of economics. I investigate the inflation dynamics using the model and estimate the instability of the Phillips curve after COVID-19. The model can also forecast atypical changes in inflation rates since COVID-19. In Chapter 2, co-authored with Ivana Komunjer, we propose a state space model for panel data, which is useful to analyze the non-constant influences of latent characteristics on economic outcome variables without proxy variables. We show the similarity transform of model parameters and identify the model parameters from the first and second moments of the data. We consider two specific situations: an assumption for the initial condition of a latent characteristic, and an assumption for the conditional variance of idiosyncratic shock. We obtain rank conditions for local identification from the similarity transformation relations of model parameters. Finally, in Chapter 3, co-authored with Youngjin Lee, we investigate the effect of expectations on house prices. Expectations on the future are crucial determinants of house prices, and many surveyed measures elicit these expectations. However, estimating the effect of the latent characteristic is difficult because various surveys exist, and the results depend on the type of survey used. We use dynamic factor models to derive proper indices for expectations from multiple measures, and we show how expectations have an important influence on house prices
The Repeating Book: In Search of the Creole Text in the Caribbean
No full textPh.D.In my dissertation, I argue that Caribbean thought and literature exhibit exceptional characteristics that make them unique, sharing features that make them what I call the “creole text.” Orlando Fals Borda’s Historia Doble de la Costa (1979-86) and Antonio Benítez Rojo's La Isla que se Repite (1989) guide my analysis. Historia Doble de la Costa is a much lesser-known book, it is a tetralogy that navigates between a poetic and a historicist discourse which, for lack of a better term, can be described as a social history whose structure duplicates its pages as if it contained two books, one of them a history from below, with parallel texts on facing pages. Historia Doble de la Costa also repeats itself because each of its four volumes repeat elements of the others to narrate the histories of the different subregional islets of the Colombian Caribbean. Through drawing on Benitez Rojo idea of “repetition”, I show that this repetition characterizes textual creolity in general, such as in C.L.R. James's The Black Jacobins (1938), Fernando Ortiz's Contrapunteo Cubano del Tabaco y el Azúcar (1940), Lydia Cabrera's El Monte (1954), George Lamming's The Pleasures of the Exile (1960), Édouard Glissant's Le Discours Antillais (1981), Manuel Zapata Olivella's Changó, el Gran Putas (1983), among others. I maintain that the creole text is cannibalistic in the sense that it is a textual assemblage that combines different fields of knowledge, approaches, intellectual traditions, formats and writing styles in an undisciplined manner. This anthropophagous attitude constitutes an artistic amalgam, unclassifiable, with a tendency towards excessiveness and the baroque, generating polyphony. In this way, the creole text opens up, including not only the oral record and myth, but also narrating the agency of subjects previously considered passive objects, in particular, the historical experience of the African diaspora
Kung Fu Nationalism: Huo Yuanjia, Wong Fei-hung, and the Making of Modern China’s Humiliation Narrative
No full textPh.D.One of the most important tales about China’s modern past is the “Century of Humiliation” (1839-1945), which begins with the first Opium War (1839-1842) and roughly concludes with Japan’s defeat during World War Two. Since the early 1990s, the Chinese Communist Party has promoted this storyline to bolster the party’s political legitimacy. But why did the Party appropriate this narrative and what explains its receptiveness in China? I argue that popular culture has played an important part in propagating the “Century of Humiliation” narrative. Drawing on Michael Billig’s theory of “banal nationalism” and highlighting the role of Chinese diaspora communities in Hong Kong, Southeast Asia, and North America, this dissertation examines the evolution of the Huo Yuanjia and Wong Fei-hung martial arts stories, as articulated in print and in film throughout the twentieth century. Huo Yuanjia (1868-1910) was a famous northern Chinese martial artist from Tianjin who gained a degree of local notoriety after he settled in Shanghai in 1909. His life story was popularized in the serialized publications of martial arts writer Xiang Kairan (1890-1957) in the 1920s and 1930s and expanded upon in the early 1970s and 1980s through Hong Kong-produced films and television miniseries. Wong Fei-hung (1847-1925) was a southern Chinese martial artist from Foshan, Guangzhou (Canton). Like Huo Yuanjia, Wong’s story initially spread through local word-of-mouth and serialized writings before being popularized in a long-running Hong Kong films series as well as widely viewed television shows. By examining the historical evolution of these stories, this research allows us look beyond a top-down, government-centered perspective that informs many studies on the history of Chinese nationalism and instead better understand the role of non-traditional actors who helped create China’s modern national narrative from the ground up
Trauma-Informed Practices for Teaching and Learning
No full textTrauma-informed pedagogy is a strategic pedagogical approach that prioritizes emotional resilience and inclusivity across academic environments. While it is a well-established paradigm in K-12 education, these pedagogical practices are not as well described in higher education. In this session, workshop participants will reflect on well-established trauma-informed principles, and through vignette based discussions, participants will collaboratively develop strategies to apply in their own educational contexts
Glucocorticoid Receptor Promotes Triple-Negative Breast Cancer Progression Through AIB1 Isoform-Expressing Cells
No full textPh.D.The glucocorticoid receptor (GR) signaling has been implicated in the metastatic progression of some breast cancer subtypes, yet it remains elusive in which context GR is relevant. Here, we investigated the role of glucocorticoid signaling in early-stage breast cancer (Ductal Carcinoma In Situ, DCIS) where a subpopulation of cancer cells that expresses the AIB1Δ4 isoform of the nuclear receptor coactivator AIB1 (Amplified In Breast Cancer 1) enables invasion and metastasis of surrounding cells. GR antagonists inhibited DCIS AIB1Δ4-expressing cells (DCISΔ4) from enabling DCIS parental cells in 3D cell cultures and in vivo. GR knockdown indicated that GR effects are predominantly mediated through DCISΔ4 cells. Uniquely activated or repressed glucocorticoid-regulated genes were detected more frequently in DCISΔ4 than in parental DCIS cells. In cocultured cell populations, activation of MYC regulators in DCISΔ4 cells correlated with significant changes in metabolism and cell growth pathways. GR CUT&RUN and ATAC-seq in dexamethasone treated DCISΔ4 vs parental DCIS cells revealed GR genomic engagement at GRE in both lines. However, there was more engagement of GR in DCISΔ4 cells which correlated with open chromatin regions. In parallel to these experiments, we discovered a murine homolog to AIB1Δ4, called AIB1Δ5. The AIB1Δ5 protein was characterized and the conservation of the enabling phenotype was confirmed in vivo. GR activation was also observed in mouse mammary gland carcinoma cells that expressed AIB1Δ5 when mixed with parental cells, indicating that AIB1Δ5 can be used in a syngeneic mouse model to further study enabling with systemic GR modulation. We conclude that differential effects of glucocorticoids on GR and AIB1-isoform genomic engagement result in distinct gene expression patterns in AIB1Δ4-expressing cells and enable them to control invasion and metastasis. This phenomenon is likely conserved in mouse cells, which would provide a means for evaluating glucocorticoid antagonists in vivo
Enhancing Cognitive Function in Aging: Investigating Amphetamine’s Role in Norepinephreine-Mediated Memory Improvement
No full textPh.D.Cognitive aging represents the decline in cognitive abilities as a consequence of the aging process, and it is characterized by the hypo-functionality of different cognitive domains such as memory, processing speed, attention, and executive function (1). This decline negatively impacts the daily life of elderly individuals and it can often become a pathological state; therefore, understanding the neural and biological mechanisms behind the cognitive aging process is fundamental as the average lifespan continues to rise.Although the precise mechanisms underlying the cognitive aging process remain elusive, previous studies have shown that aged Fisher 344 male rats display a marked deficit in glutamate-stimulated NE release in the Hippocampus and Cortex, brain regions critical for memory formation. Our laboratory not only confirmed these findings, but it also discovered that the psychostimulant amphetamine rescues this deficit restoring it to the levels seen in the young rats.
In this dissertation, I investigated the potential therapeutic effect of amphetamine in age-related cognitive decline. My data from behavioral studies demonstrated that acute and chronic amphetamine treatments enhance spatial and recognition memory in aged rats. Furthermore, the application of the non-selective-beta-adrenergic blocker propranolol abolishes amphetamine effect, indicating that the enhancing cognitive effects of amphetamine on recognition memory are mediated primarily by NE. In addition to its effects on release and cognitive function, chronic amphetamine increases the density of dendritic spines in the Cortex and Hippocampus of aged rats and shifts the proportion of spine types from immature spines (long thin) towards more mature spines (mushroom). My findings suggest that the utilization of amphetamine during normal aging could possibly mitigate age-related cognitive decline
The Relationship Between Student Disability Types and Inclusive Education Outcomes
No full textM.P.P.This paper examines how various types of disabilities among students in an inclusive education setting in the U.S. impact their high school graduation. Utilizing data from the Office of Special Education Programs through a fixed effects regression model, the study finds that deaf and blind students exhibit significantly higher dropout rates. Conversely, students with hearing impairments tend to receive alternative credentials. These insights accentuate the necessity for diverse educational policies to facilitate the success of all students, advocating more tailored approaches to inclusive education to optimize academic achievement for students with different disabilities
Random Utilities and How to Find Them
No full textPh.D.In this dissertation, I study the random utility model. The random utility model is an extension of the classic paradigm of economics which assumes that decision makers choose according to some underlying preference. The random utility model extends this paradigm by allowing for heterogeneity across either a population of decision makers or across time for the same decision maker. This heterogeneity is modeled as there being a distribution over preferences inducing a distribution over choices. In Chapter 1, I study when an analyst is able to recover the underlying distribution over preferences from choice data. I provide fully characteristic conditions under which we are able to recover the underlying distribution over preferences. In Chapter 2, I readdress the problem of testing the random utility model. While axiomatic tests of the random utility model have been known, only recently has a hypothesis test for the random utility model been developed which can be applied to real data. However, this hypothesis test is not computationally feasible in many reasonable applications. I provide an alternative hypothesis test, applicable to real data, that offers large computational improvements over the current standard methodology. In Chapter 3, I study the random utility model in a dynamic setting where a decision maker's past choices can impact their preference today. First, I broach the problem of aggregation. In general, if a decision maker's preference depends on their history of choices, the time average of their choices does not coincide with the random utility model. I provide characteristic conditions for when the random utility model is an accurate model of time aggregated choice. Second, I develop a test for this type of dynamic random utility when we have time disaggregated but population level data. I provide a fully characteristic axiomatic test as well as a hypothesis test for history dependent random utility for this type of data