Econometrics Commons^™

Forecasting The Unpredictable: The Bullwhip Effect In The Toy Industry, Mannat Sethi

CMC Senior Theses

This thesis investigates the bullwhip effect, the amplification of demand variability as it moves upstream in a supply chain, with a focus on the toy industry, and comparative analysis with the technology and consumer goods sectors. Using annual data from 1993 to 2024, the study analyzes production and retail growth for Mattel, Hasbro, Intel, and Procter & Gamble to assess how variance between consumer demand and firm output has changed over time. Bullwhip ratios are calculated as the variance of production growth divided by the variance of retail growth across two periods: 1993–2008 and 2009–2024.

The findings show that toy …

Macroeconomic Resilience And Sentiment Dynamics In The Global Luxury Fashion Industry, Mehr Gupta

CMC Senior Theses

The global luxury fashion industry is often viewed as a sector where broad economic conditions intersect with individual spending behavior. Yet despite this common assumption, it remains unclear whether macroeconomic fluctuations meaningfully shape luxury firms’ financial performance. This thesis examines the relationship between four key macroeconomic indicators - consumer sentiment, disposable income, unemployment, and exchange rates - and firm-level quarterly revenue for twelve publicly listed luxury companies from the early 2000s through 2024. Firm financial data from S&P Capital IQ are matched with macroeconomic measures from the Federal Reserve (FRED), and the analysis employs a two-way fixed-effects framework with firm …

Factors Correlated With State-Level Variation In Grocery Consumption Changes During The Covid-19 Pandemic, Sun Young Byun

CMC Senior Theses

In this paper, I employ a two-part empirical strategy to explore the effects of the COVID-19 pandemic on grocery consumption across U.S. states. First, I use the United States Department of Agriculture's Weekly Retail Food Sales data from 2018-2023 to estimate a time-series panel model. I use the results to test three hypotheses regarding state-level variation in consumption changes and impacts in the initial pandemic period, the remaining pandemic period, and the post-pandemic period. Second, I use the estimated state-level effects as dependent variables in cross-sectional regressions to test three hypotheses linking variation in state-level consumption patterns to economic, demographic, …

From Static Λ To Moving Penalties: Dynamic Regularization In Mean-Variance Portfolio Choice, Ariel Santiago

CMC Senior Theses

This thesis evaluates whether dynamically tuned weight penalties can improve high-dimensional mean-variance portfolio choice. Classical Markowitz portfolios perform poorly when the number of assets is large relative to the data, because small estimation errors in means and covariances can generate extreme, fragile allocations. A common remedy is to regularize the weights using L1 (lasso), L2 (ridge), or elastic-net penalties, but these are typically chosen once and held fixed. Using monthly excess returns on the 48 Fama-French industry portfolios from January 2010 to June 2025, this thesis compares penalized mean-variance and minimum-variance strategies under long-only and long-short constraints, with and without …

Algorithmic Trading In Idiosyncratic-Payoff Markets: A Multi-Agent System For On-Chain Prediction Contracts, Saif Aldeen A.K. Agha

CMC Senior Theses

This thesis documents the design, deployment, and forward-test evaluation of an evolutionary multi-agent algorithmic trading system on Polymarket, the largest decentralized prediction market. The system pairs a locally-hosted 72-billion-parameter language model with a gradient-boosted statistical filter and an evolutionary selection mechanism that maintains a population of approximately 500 autonomous trading agents. Each agent generates a probability estimate for an event, compares it to the prevailing market price, and trades the resulting disagreement.

The central empirical exercise estimates a panel regression of trade-level profit on the absolute disagreement between the agent's probability estimate and the market price, controlling for agent identity, …

Volatility Modeling With An Application To Risk Parity Portfolios, Kenneth Hou

CMC Senior Theses

This thesis studies volatility modeling in the context of risk parity portfolio construction. I compare three risk parity portfolios that differ only in their underlying volatility model: a historical covariance baseline, a Bayesian stochastic volatility model, and a GRU–GARCH hybrid neural network. Using daily returns on Kenneth French’s five industry portfolios from January 2016 through December 2025, I construct monthly rebalanced portfolios under each model, with the SV and GRU forecasts embedded in hybrid covariance matrices that combine forecasted volatilities with rolling historical correlations. The results document a divergence between forecast accuracy and portfolio performance: the SV model is the …

The Zoom Town Effect Of Remote Work On Rural Housing Markets, Luc R. Fabing

CMC Senior Theses

I study the impact of remote work on rural housing markets, highlighting the moderating role of outdoor recreation amenities. Using panel data from the American Community Survey and Zillow (2010–2024), I employ a Two-Way Fixed Effects model and a shift-share instrumental variable strategy to estimate the effect of work-from-home (WFH) rates on home prices. I find that higher WFH rates are strongly associated with higher home prices in high-recreation counties, but that the association is weak or negative in counties lacking significant recreation amenities. These findings point to amenity migration as a key mechanism through which remote work impacts rural …

From Collective Punishment To Collective Action, Samuel N. Johnson-Saeger

CMC Senior Theses

Regime change may be sought in two primary ways, absent a turn to internal or external violence: through nonviolent resistance, from within, and sanctions, from abroad. This paper is about how those two pathways are intertwined. I challenge two different strands of the existing literature. First, I show that Liou et al. (2022)'s finding that sanctions increase the predicted probability of nonviolent resistance movement progress is subject to substantial upward bias from country heterogeneity. Incorporating country fixed effects, I show that sanctions, on average, have no statistically significant effect on resistance outcomes. Second, I challenge Chenoweth and Stephan (2008)'s argument …

Trading Volume And Forecast Accuracy In Kalshi Prediction Markets, Nicholas Smole

CMC Senior Theses

This thesis examines the accuracy of Kalshi prediction markets and how different measures of trading volume shape their calibration and predictive power. Expanding upon previous models, use nonparametric calibration curves, pseudo-likelihood regressions, and Brier-score regressions on a large panel of Kalshi contracts to compare the effects of ex-post total volume and cumulative trading volume on market accuracy. Markets are typically well-calibrated near expiration; however, at longer horizons, low-volume markets exhibit substantially more miscalibration. Holding time to expiration fixed, greater cumulative volume is associated with higher accuracy. In comparison, higher ex-post total volume is associated with lower accuracy, highlighting the importance …

Determinants Of Remote Suitability Of Jobs And Residential Distance From The Workplace After The Covid-19 Pandemic, Abizer Mamnoon

CMC Senior Theses

Immediately before the COVID-19 pandemic, workers performed 14% of their paid hours from home, but this figure increased to 38% after the pandemic (Chen et al. 2023). This thesis contributes to our understanding of the determinants of remote work and the factors affecting how far an individual lives from the workplace in the post-pandemic period.

I use 1.95 million observations of worker-specific LinkedIn data from August 2023. I construct a measure of remote suitability using Dingel and Neiman’s (2020) method to assign each job title a remote suitability score based on the associated occupation. I use a Tobit regression to …

The Solar Power Of Pricing: Does Electricity Pricing Policy Drive Residential Solar Demand?, Talia De Beer

CMC Senior Theses

Solar energy has emerged as a pivotal renewable energy source, experiencing significant growth in the United States over the past two decades. As policymakers continue to develop and refine the best methods to incentivize solar adoption while reflecting true costs and potential equity concerns, this paper seeks to evaluate the impact of changes in solar pricing policies on residential solar installations. Employing a difference-in-differences regression model, this study examines the effects of policy changes on both the system capacity and number of installations of adoption in three Western states that have transitioned away from traditional Net Energy Metering (NEM) policies–Arizona, …

China’S Economic Slowdown: How Does It Affect New Zealand’S Economy?, Haven Qin

CMC Senior Theses

This paper examines the impact of China’s economic slowdown on New Zealand’s economy, focusing on sector-specific linkages across real estate, the stock market, exports, and import prices. As a small open economy, New Zealand’s deep integration with global markets makes it particularly vulnerable to external shocks from its largest trading partner. Employing OLS regression models and monthly economic data, this study reveals the nuanced pathways through which China’s economic conditions influence New Zealand. Key findings include the sensitivity of New Zealand’s housing market to Chinese property investments and Chinese stock market performance, the influence of Chinese financial market performance on …

When Consumers See It Coming: How Anticipation Affects The Domestic Cost Of Tariffs, Pierce D. Sanderson

CMC Senior Theses

Tariffs implemented by President Trump in 2018 were unexpected and significant, incurring large costs to U.S. consumers. This study examines the effect of tariff anticipation on price, quantity, and value effect on imports in the United States. Specifically, it focuses on tariffs implemented during Obama’s second term in office from 2012 to 2016 on the automotive, technology, and manufacturing industries. The findings present a significant divergence from the literature on the effects of 2018 tariffs, finding a near full pass-through of tariffs to foreign exporter prices rather than U.S. consumers. Specifically, a one percent increase in tariffs resulted in a …

Batteries For Grid Stability: The Impact Of Utility-Scale Battery Installation On Electricity Price Volatility In California, Hannah Vestring

CMC Senior Theses

As renewable energy penetration increases, driven largely by solar, electricity prices have become increasingly volatile due to the intermittency of supply, increasing the need for price stabilizing interventions. This study examines the impact of cumulative additions of utility-scale battery energy storage systems (BESS) capacity on real-time market (RTM) electricity price volatility within the California Independent System Operator (CAISO) between 2020 and 2024, taking a long-term, system-wide perspective on storage deployment. Using an Ordinary Least Squares (OLS) regression framework, the analysis finds that cumulative BESS capacity is significantly associated with reduced price volatility in the benchmark model, with results remaining robust …

Clutch Or Comfortable? How Franchise Tags Influence Qb Performance In The Nfl, Justin Edwards

CMC Senior Theses

Abstract

The National Football League (NFL) has a type of contract known as the Franchise Tag. The Franchise Tag is a one-year contractual extension often used when a player and a team cannot come to terms regarding the details of a contract extension. Usually, the player expects to be paid more than the organization can afford that year due to restrictions imposed by the salary cap. Because of this, Franchise-Tagged players tend to have a greater incentive to perform well in hopes of being signed into a long-term contract. My findings show that the incentive of future earnings and job …

Electric Vehicles: Exploring The Granularity In California Zip Codes, Hans Melvern Kasim

CMC Senior Theses

This paper studies a number of factors that affect electric vehicle (EV) sales in California. In comparison to previous papers, this paper examines EV sales at a higher granularity by using zip code level data. My model utilizes a panel fixed effect to examine factors like income, energy prices, commute time, number of chargers and tax rebates. The results show that the number of chargers (specifically DC fast), commute time, and tax rebates have the greatest statistically significant effects on EV sales in California. Furthermore, I utilize an empirical specification - a conditional logit model accounting for zip code fixed …

Divided By Design? Urban Renewal’S Differential Impacts On Economic Outcomes By Race, Derik Suria

CMC Senior Theses

I study city-level effects of the federally and state-sponsored urban renewal program that aimed to improve living conditions for residents in blighted areas and slums. I use an interdisciplinary approach to estimate urban renewal effects on measures of income, property value, poverty, and employment. I first replicate the methodology and estimates of urban renewal effects on city outcomes in 1980 from Slum Clearance and Urban Renewal in the United States (Collins and Shester 2013). I extend this research by looking at a longer time horizon (1990, 2000, 2010) and incorporating race-based outcomes. From 1980 to 2010, estimated effects on property …

Driven By Change: The Impact Of Macroeconomic Shifts And Covid-19 On New Vehicle Sales, Jackson Aldrich

CMC Senior Theses

This paper examines the impact of macroeconomic factors and the COVID-19 pandemic on new vehicle sales. In order to address these two topics, a two-pronged approach was used with separate regression models. The macroeconomic variables include monthly supply of new homes, CPI for urban public transportation, unemployment rate, disposable personal income, inflation expectation, consumer sentiment, average gas prices, and total vehicle miles traveled which were regressed on total vehicle sales from 1978-2022. The regression results confirmed and supported current literature and highlighted the importance of the housing market and unemployment rate on new vehicle sales. The COVID-19 pandemic model variables …

Determinants And Effects Of Offset Usage Within The California Cap-And-Trade System – A Firm-Level Analysis, Adarsh Srinivasan

CMC Senior Theses

This thesis examines the determinants of the usage of carbon offsets within the California cap-and-trade program for greenhouse gas emissions. It also investigates the relationship between a firm’s investment in offsets and its financials for publicly traded companies. Entity-level data from 2013 to 2020 was used, which was sourced from the California Air Resources Board. Multiple regression analysis was used to investigate factors that correlated with offset usage and whether offset usage correlated with EBITDA. Dummy variables indicating geographic location and economic sector of each entity were used as controls. The analysis found that total covered emissions and privately owned …

Guns And Fatal Police Shootings: Accuracy Of Firearm Prevalence Proxies In A Panel Data Analysis, Corin Elmore

CMC Senior Theses

This paper explores the relationship between gun prevalence and fatal police shootings in the United States. Specifically, the study assesses the validity of using FSS (Firearm Suicides Per Total Suicides) as a proxy for gun prevalence. It examines whether FBI background check data can be a more reliable alternative. Through a panel data analysis, the study provides moderate evidence that FBI background check data performs better than FSS as a proxy for gun prevalence and that the relationship between firearm prevalence and fatal police shootings is positive. Additionally, the study finds that instrumenting FSS with background check data to account …

Metrics At 8,000 Meters: Determinants Of Summit Success In The Himalayas, Alyssa Gallagher

CMC Senior Theses

This paper examines the determinants of success for individuals summiting 8,000-meter Himalayan peaks from 1995 to 2019. Using binary logistic regressions, this paper presents evidence that a climber’s citizenship is a highly significant determinant of success. Chinese citizens are clear outliers with 192% more odds of success than an American. The success rates related to season, gender, and year all conform with existing research, justifying the robustness of the citizenship trends. Despite controlling for per capita income, the results assert that wealth may not be an important determinant of success and provokes interesting questions as to what factors may influence …

Impact Of Recessions And The Business Cycle On Altruism, Nicolas De Mello

CMC Senior Theses

I examine how recessions and the business cycle impact the time individuals allocate to altruism, using American Time Use Survey data from 2002-2019. The aim of the paper was to create a measure of altruism to explore changes in attitudes towards charity and altruism instead of the financial ability to do so. Additionally, I want to understand how economic shocks, such as recessions, impact behaviors and habits across the US population. Using ordinary least squares and the Tobit model, I created two models to capture both changes during and after recessionary periods. The first model is a regression of individuals’ …

Renewable Portfolio Standards: Effectiveness And Carbon Implications, Alexander S. Albrecht

CMC Senior Theses

A renewable portfolio standard (RPS) policy is a popular regulatory tool implemented within the U.S. and abroad to limit energy sector emissions and incentivize renewable energy. Assessing their effectiveness and efficiency is a key component of achieving further reductions. We assess an energy market under an RPS using fixed-effects panel and 2SLS regression models to lend empirical credence to common theory-based concerns about RPS policy, namely (1) that they leave emissions unregulated once the RPS requirement is met and (2) that they do not incentivize full use of renewable energy resources. Our results show these to be valid concerns that …

The Paycheck Protection Program's Effect On Real Estate Prices, Someswar Amujala

CMC Senior Theses

This paper investigates the relationship between the Paycheck Protec- tion Program (PPP) and real estate price increases. The Paycheck Protection Program was established by the Small Business Association in 2020 to pro- vide forgivable loans to businesses to aid with potential losses from COVID- 19 impacts. I leverage zip code level data across the United States and a fixed effects panel data model to quantitatively measure the PPP’s influence on housing and rental prices. I find a positive and significant relationship between the number of PPP loans disbursed and housing and rental rates. Specifically, a 1% change in the number …

Impact Of Demand For H1b Visas In The Us On The Performance Of Listed Indian Firms, Jayaditya Shekhar Maliye

CMC Senior Theses

This thesis examines the relationship between Indian firm performance and the demand for H1B visas in the US. To test this relationship, a fixed effects model has been used on a panel dataset of 104 Indian firms listed on the Bombay Stock Exchange (BSE) and National Stock Exchange of India (NSE) over the period 2009 to 2021. The results show that an increase in H1B visa demand is negatively and significantly associated with Indian firm performance. These findings suggest that the US’s demand for high skilled labor overseas has consequences for firms in the home country.

Application Of Sentiment Analysis And Machine Learning Techniques To Predict Daily Cryptocurrency Price Returns, Edward Wu

CMC Senior Theses

This paper examines the effects of social media sentiment relating to Bitcoin on the daily price returns of Bitcoin and other popular cryptocurrencies by utilizing sentiment analysis and machine learning techniques to predict daily price returns. Many investors think that social media sentiment affects cryptocurrency prices. However, the results of this paper find that social media sentiment relating to Bitcoin does not add significant predictive value to forecasting daily price returns for each of the six cryptocurrencies used for analysis and that machine learning models that do not assume linearity between the current day price return and previous daily price …

Finding Answers In Gaps: The Relationship Between Drugs And Mexico's Economy, Mariela Centeno

CMC Senior Theses

This paper sets out to understand the economic conditions that paved the way for the evolution and resilience of a powerful and resilient drug economy in Mexico. To do this, I examine the relationship between a proxy for drug movement from Mexico and large economic variables in the country. I use a dataset that incorporates the drug seizures from 1997 to 2022 to understand whether privatization and the resulting rising economic inequalities in Mexico paved the way for a growing drug economy. This paper builds on current literature regarding drug market estimates and drug economy interactions. Using an OLS regression …

Feature Investigation For Stock Returns Prediction Using Xgboost And Deep Learning Sentiment Classification, Seungho (Samuel) Lee

CMC Senior Theses

This paper attempts to quantify predictive power of social media sentiment and financial data in stock prediction by utilizing a comprehensive set of stock-related fundamental and technical variables and social media sentiments. For conducting sentiment analysis, this study employs a pretrained finBERT model that provides three different sentiment classifications and respective softmax scores. Hence, the significance of these variables is evaluated with XGBoost regression and Shapley Additive exPlanations (SHAP) frameworks. Through investigating feature importance, this study finds that statistical properties of sentiment variables provide a stronger predictive power than a weighted sentiment score and that it is possible to quantify …

An Evaluation Of Knot Placement Strategies For Spline Regression, William Klein

CMC Senior Theses

Regression splines have an established value for producing quality fit at a relatively low-degree polynomial. This paper explores the implications of adopting new methods for knot selection in tandem with established methodology from the current literature. Structural features of generated datasets, as well as residuals collected from sequential iterative models are used to augment the equidistant knot selection process. From analyzing a simulated dataset and an application onto the Racial Animus dataset, I find that a B-spline basis paired with equally-spaced knots remains the best choice when data are evenly distributed, even when structural features of a dataset are known …

Determinants Of Housing Supply Expansion In The Western United States, Nathaniel Tolles

CMC Senior Theses

New residential construction is an important indicator of economic health. Previous empirical work demonstrates the profound power of housing starts in forecasting recession. Theoretical research, backed by empirical study, suggests that home prices and interest rates are closely related to the amount of residential investment. This paper attempts to better understand the complex relationship between various factors that influence the supply and demand of new housing; what information do suppliers and regulators use to determine how many new units of housing will be constructed? Specifically, we will look at the respective state housing markets of California, Oregon, and Washington by …