Sloan School of Management
http://hdl.handle.net/1721.1/1777
2017-10-17T12:56:09ZEssays in financial economics
http://hdl.handle.net/1721.1/108220
Essays in financial economics
Sun, Yang, Ph. D. Massachusetts Institute of Technology
This thesis consists of three essays in corporate finance and capital markets. The first chapter estimates the effect of competition from low-cost index funds on fees in the money management industry. A difference-in-differences analysis exploiting the staggered entry of index funds finds that while actively managed funds sold directly to retail investors reduce fees by six percent, those sold through brokers increase fees by four percent. Additionally, actively managed funds, especially closet indexers, shift away from holding the index portfolio. The paper proposes a price-discrimination model to illustrate that the effect of low-cost passive fund competition depends on market segmentation. Beyond the price competition effect, the entry creates a selection effect that isolates the least-price-sensitive investors in the broker channel and results in a price increase for this group. Repeating the study using the entry of exchange-traded funds reveals similar but stronger finding. Overall, the results shed light on why aggregate mutual fund fees decline slowly despite increased competition from lower-cost passive alternatives. The second chapter, joint with Jean-Noel Barrot, examines the effects of imperfect investor risk adjustment on the behavior of mutual fund managers. We exploit a natural experiment when a major fund rating company changed its rating methodology. While in the old system, all equity funds were compared with one another in one pool, in the new algorithm, funds become compared within narrow peer groups. This algorithm revision increases the ability of retail investor to compare funds based on risk-adjusted returns, and it has an important impact on the fund mangers' compensation. The sensitivity of retail fund flows to systematic returns is eliminated. Using institutional funds as a control for retail funds in a difference-in-differences analysis, we find that this revision reduces fund managers risk taking behavior, in particular for funds in the categories that had biased low ratings ex-ante. The third chapter, joint with Carola Frydman and Eric Hilt, documents the dividend policy of firms in the United States during the first three decades of the twentieth century. This period features severe information asymmetry between insiders and outsiders, while other factors that could affect the payout policy were relatively muted. In the years surrounding World War I, industrial firms increased their payout ratios and dividends became less sticky. The new industrial firms listed on the NYSE in the 1920s had the best fit with the Lintner (1956) model and these firms refrained from committing to sticky dividend policy. Consistent with the asymmetric information theory, the market reacted positively to dividend increase announcements, especially to those made by the new industrials, and reacted negatively to dividend cuts.
Thesis: Ph. D., Massachusetts Institute of Technology, Sloan School of Management, 2015.; Cataloged from PDF version of thesis.; Includes bibliographical references (pages 157-163).
2015-01-01T00:00:00ZHigh dimensional revenue management
http://hdl.handle.net/1721.1/108211
High dimensional revenue management
Ciocan, Dragos Florin
We present potential solutions to several problems that arise in making revenue management (RM) practical for online advertising and related modern applications. Principally, RM solutions for these problems must contend with (i) highly volatile demand processes that are hard to forecast, and (ii) massive scale that makes even basic optimization problems challenging. Our solutions to these problems are interesting in their own right in the areas of stochastic optimization, high dimensional learning and distributed optimization. In the first part of the thesis, we propose a model predictive control approach to combat volatile demand. This approach is conceptually simple, uses available demand data in a natural way, and, most importantly, can be shown to generate significant revenue advantages on real-world data from ad networks. Under mild restrictions, we prove that our algorithm achieves uniform relative performance guarantees vis-a-vis a clairvoyant in the face of arbitrary volatility, while simultaneously being optimal in the event that volatility is negligible. This is the first result of its kind for model predictive control. While our approach above is effective at hedging demand shocks that occur over "large" time horizons, it relies on the ability to estimate snapshots of the prevailing demand distribution over "short" time horizons. The second part of the thesis deals with learning the extremely high dimensional demand distributions that are typical in display advertising applications. This work exploits the special structure of the display advertising version of the NRM problem to achieve a sample complexity that scales gracefully in the dimensions of the problem. The third part of the thesis focuses on the problem of solving terabyte sized LPs on an hourly basis given a distributed computational infrastructure; solving these massive LPs is the computational primitive required to make our model predictive control approach practical. Here we design a linear optimization algorithm that fits a paradigm for distributed computation referred to as 'Map-Reduce'. An implementation of our solver in a shared memory environment where we can benchmark against solvers such as CPLEX shows that the algorithm outperforms those solvers on the types of LPs that an ad network would have to solve in practice.
Thesis: Ph. D., Massachusetts Institute of Technology, Sloan School of Management, 2014.; Cataloged from PDF version of thesis.; Includes bibliographical references (pages 149-153).
2014-01-01T00:00:00ZOptimization driven approaches for subsidy allocation and supply chain procurement
http://hdl.handle.net/1721.1/108210
Optimization driven approaches for subsidy allocation and supply chain procurement
Romero, Gonzalo (Gonzalo Ignacio Romero Yáñez)
This thesis introduces several new models in operations management, that are motivated by practical settings. It studies these models in an optimization-driven approach, employing mathematical programming techniques to derive important structural and algorithmic insights on the corresponding problems. In the first part of the thesis, we study subsidy allocation problems under budget constraints and endogenous market response, where the central planner's objective is to maximize the market consumption of a good. We first consider co-payment subsidies, that are paid to manufacturing firms per unit sold. We focus on "uniform co-payments", in which each firm receives the same co-payment, regardless of its cost structure, or efficiency. Uniform co-payments are frequently implemented in practice. Therefore, a natural question is whether uniform co-payments are in fact the best that the central planner can do; or, more generally, how do they perform compared to the optimal co-payment allocation? Notably, we first identify relatively general sufficient conditions such that uniform co-payments are optimal, even if the firms are heterogeneous, and if the central planner is uncertain about the market response. We then complement the effectiveness of uniform co-payments, by studying a very relevant setting where they are not optimal. We show that, for any instance of this model, uniform co-payments are guaranteed to induce at least 85% of the optimal market consumption. In summary, uniform co-payments turn out to be surprisingly powerful in maximizing the market consumption of a good. We then consider lump sum subsides, which are an alternative subsidy mechanism also implemented in practice. We show that the problem of optimally allocating lump sum subsidies is NP-hard, and discuss two simple allocation policies that have good performance guarantees. In the second part of the thesis, we introduce a model to incorporate the cost of handling orders at a central distribution center, into the procurement decisions of a company. We show how structural results for this model lead to a practical method to select the best case pack size per SKU in procurement contracts, as well as to serve orders at the distribution center. Furthermore, we test this method on real data from a large utility company, finding significant total cost reductions.
Thesis: Ph. D., Massachusetts Institute of Technology, Sloan School of Management, 2014.; Cataloged from PDF version of thesis.; Includes bibliographical references (pages 179-184).
2014-01-01T00:00:00ZThe effect of the dual-career marriage on female managers in two large multistate companies
http://hdl.handle.net/1721.1/108208
The effect of the dual-career marriage on female managers in two large multistate companies
Hale, Roger Warner; Drummond, Jere Alan
Thesis. 1979. M.S.--Massachusetts Institute of Technology. Alfred P. Sloan School of Management.; MICROFICHE COPY AVAILABLE IN ARCHIVES AND DEWEY.; Bibliography: leaves 109-111.
1979-01-01T00:00:00Z