Abstract
dc:description.abstractThe widespread collection of data from mobile and wearable devices has created unprecedented opportunities to study human behavior in fine temporal resolution. One common structure for such data is categorical sequences—ordered, multinomial observations across many time points. These sequences present unique statistical challenges due to their high dimensionality and complex short- and long-term temporal dependence. The goal of this thesis is to further develop statistical tools motivated by real-world questions on such categorical sequences. We begin by addressing the problem of synthesizing sequential categorical data. The ability to synthesize realistic data in a parametrizable way is valuable for a number of reasons, including privacy, missing data imputation, and evaluating the performance of statistical and computational methods. When the underlying data generating process is complex, data synthesis requires approaches that balance realism and simplicity. We propose the paired Markov Chain (paired-MC) method, a flexible framework that produces sequences that closely mimic real data while providing a straightforward mechanism for modifying characteristics of the synthesized sequences. We demonstrate the paired-MC method on two datasets, one reflecting daily human activity (time use) patterns collected via a smartphone application, and one encoding the intensities of physical activity measured by wearable accelerometers. In both settings, sequences synthesized by paired-MC better capture key characteristics of the real data than alternative approaches. We then move to the realistic question of dimension reduction. There has been relatively little methodological development focusing on principled dimension reduction specifically tailored to categorical sequences. We aim to develop and evaluate approaches to identifying "key" sequence positions which distinguish sequence types. We frame this challenge as a regression problem, introduce a variety of regularization techniques that could be applied to achieve position-based dimension reduction, and evaluate them on the motivating dataset that reflects daily time use patterns collected via a smartphone application. Results show that the thresholded LASSO, a relatively underused technique, performs better than more established methods for data with complex sequential structure. Finally, we consider prediction tasks arising in settings such as online prediction, where only partially observed categorical sequences are available and the goal is to infer behavior in the unobserved portion of the sequence. In this work, we frame prediction in categorical sequences from a missing-data perspective and address prediction tasks through sequence synthesis. This perspective allows both time-point and structural prediction tasks to be handled within a common framework while naturally incorporating uncertainty through repeated synthesis. We investigate the behavior of synthesis-based methods across a range of prediction tasks and times when data become unobserved. Results uncover the relationships between prediction accuracy and multiple factors affecting performance, including certain intrinsic characteristics of sequences.
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Huang, Zuofu
Subjects
dc:subject × 3Rights
- Language dc:language.iso
- en
Identifiers
dc:identifier.*- Dc Identifier Citation
- Huang, Zuofu. (2026). Statistical Methods in Sequential Categorical Data. Retrieved from the University Digital Conservancy, https://hdl.handle.net/11299/282371.