Firm
Impatient Buyers and e-commerce Keyword Bidding
In today’s fast-paced e-commerce landscape, buyer impatience plays a crucial role in shaping the competitive dynamics between sellers and platforms. This project examines how characteristics of buyers, products, and market structure influence the equilibrium bid price in keyword auctions conducted by platforms like Google and Amazon. Focusing on the implications of buyer impatience, we explore how the value of top search result positions may increase and whether platforms can exploit this behavior for additional profits.
Online Dating Project
In collaboration with an online dating app company, we evaluate detailed user data on the two-sided matching market. Our projects include (1) estimating a spatial learning model, where the dating app users sequentially update their beliefs on match probabilities, (2) building a dashboard for the company to manage several experiments at once and to view statistically robust results, and (3) developing machine learning methods to rate the attractiveness of users’ photos.
Admin Data
Personalized learning system
In this line of research, we work together with the educational platform PaGamO to create a workable e-learning system. This study aims to enhance rural kids’ academic performance and support teachers in assessing their students’ progress. To this end, we apply the Rasch model, which interprets students’ performance as a function of students’ abilities and difficulties of items, to analyze the data. Under this structure, we estimate the difficulties of items and the trajectories of students’ abilities. Eventually, we hope to develop an algorithm based on abilities and difficulties to facilitate students’ learning performance.
Health Outcome and Choice of Caregiver
The purpose of this study is to investigate how the health outcomes of older adults are influenced by their choice of caregivers. Using data from the Taiwan Longitudinal Study on Aging (TLSA), we analyze the relationship between caregivers and older adults’ health outcomes. Specifically, we describe the choice of caregivers and older adults’ respective health outcomes for the caregivers. We also employ the instrument of caregiver reform to estimate the potential changes in health outcomes. This study provides insights into the impact of caregivers on the health of older adults, which can inform policies and interventions aimed at improving the health and well-being of older adults.
Can Immigration Limit Native-Born Demographic Decline? Old-Age Health Shocks, Migrant Care-Giving, and the Growth of Families
Taken together, falling fertility rates and rising lifespans are causing many of the worlds wealthiest countries to face steep increases in age-related long-term care (LTC) needs that must be born by a comparatively shrinking native labor force. This paper exploits rich administrative records to provide novel evidence on the scope and distribution of those costs. Beyond their well-studied disemployment effects, we find that health shocks to the elderly cause large families to grow larger. adult children are more likely to get married, and they have more children of their own. These responses are persistent over time, and they are consistent with the idea that LTC needs induce family members to substitute from formal employment into a mix of informal caregiving and home production. Members of smaller families, on the other hand, experience sharp increases in mortality risk, which is consistent with caregiving-related deaths of despair.” Leveraging quasi-experimental variation in access to international caregiver markets for individual families, we find that formal caregiver hiring amplifies positive fertility responses, whereas mortality responses vanish. These results suggest that immigration policy can causally improve native-born well-being and sustain longer-term native-born demographics.
Effects of House-Inheriting on Labor and Family Decisions
People’s response to inheritances has been a crucial topic in public finance literature. This research estimates the causal effects of inheriting a house on work, marital, and fertility outcomes. With the Taiwanese administrative tax data, we use an event study design to compare those who suffered from the loss of parents and received house inheritances to those who did not. We find that inheriting houses from parents decreases the labor supply and increases marriage and fertility rates. The significant increases in marriage and fertility support the long conjectured anecdote that house possession plays a crucial role in household formation in Taiwanese culture. We also observe rich heterogeneity in age and gender, where young inheritance recipients display more significant effects. Similar effects are also found in spouses married to recipients.
Others
Two Step Empirical Bayes
The research project focuses on the estimation of parameters in a non-linear y model, where the observables x and y are determined by an unobserved variable theta generated by a prior. In such cases, the traditional method of using the posterior mean of theta to run regression against y, which is accurate in linear models, may result in biased estimation. To address this issue, we propose a new method with its statistical properties and simulation examples. This project aims to improve parameter estimation in non-linear models under Empirical Bayes context and compare the performance with non-parametric methods.
Empirical Bayes with Trees
This project proposes an empirical Bayes method, called “EB with Optimal Shrinkage Trees”, for estimating treatment effects in settings with many treatment arms and moderate sample sizes. It leverages auxiliary information from treatment characteristics by grouping similar treatment arms using a decision tree, then shrinking individual effect estimates towards the group average. The proposed method reduces mean squared errors compared to methods with no shrinkage or conventional EB methods that do not consider treatment characteristics. A consistent model selection procedure approximates the optimal tree, and simulations show reduced estimation errors, particularly when treatment characteristics highly correlate with treatment effects.
Is being handsome great? Using dating apps as evidence.
This project aims to investigate the correlation between facial attractiveness and human behavior, particularly in the context of online dating. By using machine learning algorithms to predict facial beauty, we will explore the marginal utility of attractiveness in the online dating market. Our study will also examine how physical appearance affects human behavior and decision-making in the dating process. With a large dataset of profile pictures from various dating apps, we will train our model to accurately classify images based on their level of facial attractiveness. Through this study, we hope to provide insights on the significance of physical appearance in human behavior and decision-making, especially in the context of online dating.