Shrinking the Cross Section∗

By S. Kozak, Stefan Nagel, S. Santosh. 2017

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Deep Learning for Predicting Asset Returns (2018, methodology): “Improved methodologies have been suggested by Campbell and Thompson (2007), Rapach et al. (2010) and Harvey et al. (2016) and, more recently, Feng et al. (2017) and Kozak et al. (2017).”
Taming the Factor Zoo: A Test of New Factors (2019, unknown): “An alternative, following Kozak, Nagel, and Santosh (2020), is to first construct PCA of the factors, and then use LASSO on the principal components.”
Large Dimensional Latent Factor Modeling with Missing Observations and Applications to Causal Inference (2019, unknown): “t factors which can explain well the cross-section of expected returns.31 We assume that 30A small number of anomalies have also a few missing values at the beginning of the sample. 31See for example(Kozak et al., 2019; Kelly et al., 2018; Lettau and Pelger, 2018). 31 the stochastic discount factor ”
Short- and Long-Horizon Behavioral Factors (2019, unknown): “Based on the arguments in Daniel, Hirshleifer, and Subrahmanyam (2001) and Kozak, Nagel, and Santosh (2017a), this will lead to a high return premium for firms that load on the resulting PEAD factor.”
Deep Factor Alpha (2018, background): “Deep learning differs from sparsity model (Kozak, Nagel, and Santosh (2017)) in a number of important ways.”
Deep Learning in Asset Pricing (2019, methodology): “…onto quantiles of characteristics are exactly the input to PCA in Kelly, Pruitt, and Su (2019) or the elastic net mean-variance optimization in Kozak, Nagel, and Santosh (2020).12 The solution to minimizing the sum of squared errors in these moment conditions is a simple mean-variance…”
Understanding Systematic Risk: A High‐Frequency Approach (2020, unknown): “Kozak, Nagel, and Santosh (2020) apply mean-variance optimization with an elastic net to PCA-based factors.”
Empirical Asset Pricing via Machine Learning (2018, unknown): “The second strand forecasts the time series of returns and is surveyed by Welch and Goyal (2008), Koijen and Nieuwerburgh (2011), and Rapach and Zhou (2013). This literature typically conducts time-series regressions of broad aggregate portfolio returns on a small number of macroeconomic predictor v”
A Transaction-Cost Perspective on the Multitude of Firm Characteristics (2020, unknown): “The elastic-net SDF approach in Kozak et al. (2017) shrinks the contributions to the SDF of both low-variance principal 16The term lasso originated as the acronym for least absolute shrinkage and selection operator.”