Jim Chen's SSRN abstracts

Welcome to Jim Chen's SSRN abstracts. This is a human-friendly display of the RSS feed  for Mr. Chen's official SSRN page (http://ssrn.com/author=68651), reorganized by reverse chronological order rather than number of downloads. To receive updates as Mr. Chen posts new papers or updates old papers, please use the following form:

Enter your e-mail address:    


New: The Capital Asset Pricing Model, https://privwww.ssrn.com/abstract=3844183 (May 12, 2021)

The capital asset pricing model (CAPM) is the dominant paradigm in financial risk management. It formalizes mean-variance optimization of a risky portfolio given the presence of a risk-free investment such as short-term government bonds. The CAPM defines the price of financial assets according to the premium demanded by investors for bearing risk in excess of the risk-free return.
New: Clustering Commodity Markets in Space and Time: Clarifying Returns, Volatility, and Trading Regimes Through Unsupervised Machine Learning, https://privwww.ssrn.com/abstract=3791138 (March 31, 2021)

Unsupervised machine learning can interpret logarithmic returns and conditional volatility in commodity markets. k-means and hierarchical clustering can generate a financial ontology of markets for fuels, precious and base metals, and agricultural commodities. Manifold learning methods such as multidimensional scaling (MDS) and t-distributed stochastic neighbor embedding (t-SNE) enable the visualization of comovement and other financial relationships in three dimensions.

Different methods of unsupervised learning excel at different tasks. k-means clustering based on logarithmic returns works well with MDS to classify commodities and to create a spatial ontology of commodities trading, A strikingly different application involves k-means clustering of the matrix transpose, such that conditional volatility is evaluated by trading date rather than by commodity. This approach can isolate the two most calamitous temporal regimes of the past two decades: the global financial crisis ...
New: Interpreting Linear Beta Coefficients Alongside Feature Importances in Machine Learning, https://privwww.ssrn.com/abstract=3795099 (March 29, 2021)

Machine-learning regression models lack the interpretability of their conventional linear counterparts. Tree- and forest-based models offer feature importances, a vector of probabilities indicating the impact of each predictive variable on a model’s results. This brief note describes how to interpret the beta coefficients of the corresponding linear model so that they may be compared directly to feature importances in machine learning.
REVISION: Models for Predicting Business Bankruptcies and Their Application to Banking and Financial Regulation, https://privwww.ssrn.com/abstract=3329147 (February 26, 2021)

Models for predicting business bankruptcies have evolved rapidly. Machine learning is displacing traditional statistical methodologies. Three distinct techniques for approaching the classification problem in bankruptcy prediction have emerged: single classification, hybrid classifiers, and classifier ensembles. Methodological heterogeneity through the introduction and integration of machine-learning algorithms (especially support vector machines, decision trees, and genetic algorithms) has improved the accuracy of bankruptcy prediction models. Improved natural language processing has enabled machine learning to combine textual analysis of corporate filings with evaluation of numerical data. Greater accuracy promotes external processes of banks by minimizing credit risk and by facilitating regulatory compliance.
REVISION: Split Decisions: Practical Machine Learning for Empirical Legal Scholarship, https://privwww.ssrn.com/abstract=3731307 (February 6, 2021)

Multivariable regression may be the most prevalent and useful task in social science. Empirical legal studies rely heavily on the ordinary least squares method. Conventional regression methods have attained credibility in court, but by no means do they dictate legal outcomes. Using the iconic Boston housing study as a source of price data, this Article introduces machine-learning regression methods. Although decision trees and forest ensembles lack the overt interpretability of linear regression, these methods reduce the opacity of black-box techniques by scoring the relative importance of dataset features. This Article will also address the theoretical tradeoff between bias and variance, as well as the importance of training, cross-validation, and reserving a holdout dataset for testing.
REVISION: An Introduction to Machine Learning for Panel Data, https://privwww.ssrn.com/abstract=3717879 (January 28, 2021)

Machine learning has dramatically expanded the range of tools for evaluating economic panel data. This paper applies a variety of machine-learning methods to the Boston housing dataset, an iconic proving ground for machine learning. Though machine learning often lacks the overt interpretability of linear regression, methods based on decision trees score the relative importance of dataset features. In addition to addressing the theoretical tradeoff between bias and variance, this paper discusses practices rarely followed in traditional economics: the splitting of data into training, validation, and test sets; the scaling of data; and the preference for retaining all data. The choice between traditional and machine-learning methods hinges on practical rather than mathematical considerations. In settings emphasizing interpretative clarity through the scale and sign of regression coefficients, machine learning may best play an ancillary role. Wherever predictive accuracy is paramount, ...
REVISION: Split Decisions: Decision Tree-Based Machine Learning for Empirical Legal Scholarship, https://privwww.ssrn.com/abstract=3731307 (December 11, 2020)

The most prevalent and useful task in social science may be multivariable regression. Empirical legal studies rely heavily on the ordinary least squares method. Using the iconic Boston housing study as a source of price data, this article introduces machine-learning methods based on decision trees and their ensembles as additional methods for regression. Although trees and forests lack the overt interpretability of linear regression, these methods reduce the opacity of black-box techniques by scoring the relative importance of dataset features. This paper will also address the theoretical tradeoff between bias and variance, as well as the importance of training, cross-validation, and reserving a holdout dataset for testing.
REVISION: An Introduction to Machine Learning for Panel Data: Decision Trees, Random Forests, and Other Dendrological Methods, https://privwww.ssrn.com/abstract=3717879 (December 11, 2020)

Perhaps no task is more prevalent, and more useful, in economic analysis than the prediction of a numerical value through its relationship with other variables. By far the most popular tool for regression is the multivariable generalization of ordinary least squares.

Machine learning and artificial intelligence have dramatically expanded the range of tools available in economics. Open-source software and a burgeoning coding community have made these methods more accessible to a broader audience.

"Dendrological" machine-learning methods use decision trees to divide data, variable by variable. Ensembles of decision trees harness the Delphic wisdom of potentially thousands of miniature regressors.

Trees and forests admittedly lack the overt interpretability of linear regression. Machine-learning often offset the opacity of these "black-box" techniques by scoring the relative importance of dataset features. This paper will also address the theoretical tradeoff ...
New: Forecasting Mortgage Demand: An Application of Traditional Methods, Machine Learning, and Neural Networks, https://privwww.ssrn.com/abstract=3656924 (August 27, 2020)

Demand forecasting relies heavily on traditional methods with well known limitations. Improved accuracy in predicting demand for mortgages, whether for purposes of purchase or refinance, is critical to profitability in home lending. To overcome obstacles to prediction using nonlinear relationships between variables and to long-term accuracy, we apply time-invariant machine-learning methods such as random forests. We also perform time-series analysis with a wide variety of deep learning architectures, including convolutional and recurrent neural networks. Time-series analysis through deep learning produces the most accurate results. Even shortcomings in forecast accuracy can reveal tacit changes in relationships among household-level and macroeconomic predictors of mortgage demand.
REVISION: After Agrarian Virtue, https://privwww.ssrn.com/abstract=3449558 (July 2, 2020)

What constitutes agrarian virtue? Across human or even geological history, agrarian virtue subsists in the sustained production of food, fiber, and fuel without the exhaustion of finite resources or the undue disruption of evolutionary processes on which human survival depends. Contemporary agricultural law, however, often emphasizes the expressive self-actualization of food preferences. This natural sublimation of economic independence from producers to consumers epitomizes agrarian vice. Restoration of agrarian virtue demands not telos (τέλος) in its purposive sense, but rather acceptance of kyklos (κύκλος), or cyclicality in its full economic and ecological sense.

  

free web page hit counter
Powered by Feed Informer