Date of Award


Degree Type


Degree Name


Degree Program

Financial Economics


Economics and Finance

Major Professor

Pezzo, Luca

Second Advisor

Velu, Raja

Third Advisor

Naka, Atsuyuki

Fourth Advisor

Shin, Seungho

Fifth Advisor

Hassan, Kabir


The dissertation consists of two chapters discussing the asset pricing models in Cryptocurrency markets and stock markets. In the first paper, we design an equilibrium model for the dynamics of the Crypto-currency mining markets in the presence of professional miners and casual miners. We apply our setup to the Bitcoin and Ethereum markets and find that major changes in their prices reflect substantial mining technological enhancements. Our model recovers, by applying Kalman filter, the dynamics of two critical but unobservable quantities: the supply for new computing power, which can be thought of as a random walk capturing the marginal enchantments in the mining technology, and the fixed cost of mining, which is found to decrease exponentially over time. We find a higher efficiency in the market for Bitcoins over Ethereum. Our empirical results show that when the price of Crypto-currency increases, the supply of computing power will also increase, following the direction of the price of Crypto-currency. Causal miners are crowded out, as a result, professional miners are dominant in mining market and there most likely exists monopoly in the mining markets when the price of Crypto-currency keeps increasing. In the second paper, we employ Reduced Rank Regression method to extract GLS latent factors through the time-varying asset characteristics. We show that GLS latent factors have superior efficiency compared to OLS latent factors. We demonstrate that Instrumented Principal Component Analysis (IPCA) introduced by Kelly et al. (2019) and Projected Principal component Analysis (PPCA) introduced by Fan et al. (2016) can be cast in the framework of Reduced Rank Regression that can accommodate both cross-sectional and time dependence. The RRR formulation also provides a framework for rich extensions. We develop maximum likelihood estimators, provide the asymptotic theory and study sparseness. The additional information contained in the more general covariance structure allows for a better and more parsimonious fit and unconditional spanning of the mean-variance frontier. We derive a closed-form limiting distribution for the regression parameters which allows for a more precise mispricing inference.


The University of New Orleans and its agents retain the non-exclusive license to archive and make accessible this dissertation or thesis in whole or in part in all forms of media, now or hereafter known. The author retains all other ownership rights to the copyright of the thesis or dissertation.

Available for download on Wednesday, May 31, 2028