The Three-Pass Regression Filter: A New Approach to Forecasting Using Many Predictors

By Bryan T. Kelly, Seth Pruitt. 2015

353 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Finance, Management Science and Operations Research

How researchers see this paper

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

In Economics and Econometrics

In Finance

In Management Science and Operations Research

Citations in the corpus

Sufficient Forecasting Using Factor Models (2015, background): “Now, (2.1) reduces to Such linear forecasting problems using many predictors have been studied extensively in the econometrics literature, for example, Stock and Watson (2002 a , b ), Bai and Ng (2008), and Kelly and Pruitt (2015), among others.”
Text as Data (2017, background): “23While Manela and Moreira (2015) study aggregate market volatility, Kogan et al. (2009) and Boudoukh et al.”
Forecasting with Factor-Augmented Regression: A Frequentist Model Averaging Approach, Second Version (2013, neutral): “Our comparisons are quite thorough, comparing our procedures with AIC selection, BIC selection, Mallows selection, CV selection, approximate Bayesian model averaging, equal weights, and the three-pass regression filter of Kelly and Pruitt (2012).”
A unified measure of Fed monetary policy shocks (2019, background): “3See Wold (1966, 1975) and Kelly and Pruitt (2013, 2015) for applications to equity returns. applicability.”
Macroeconomic forecast accuracy in a data‐rich environment (2019, neutral): “Details on the construction of the series can be found in McCracken and Ng (2015). The empirical exercise is easier when the data set is balanced. In practice, there is usually a trade-off between the relevance and the availability (and frequency) of a time series. Not all series are available from ”
Investor Attention and Stock Returns (2020, neutral): “By design, PCA is limited in its ability to capture the maximum information that is related to stock returns (Kelly and Pruitt (2015)).”
Bond Risk Premia with Machine Learning∗ (2018, neutral): “More recent work by Kelly and Pruitt (2013), Kelly and Pruitt (2015), Kozak, Nagel and Santosh (2017), Feng et al. (2017), Freyberger et al. (2017), Giglio and Xiu (2017), Feng, Polson and Xu (2018), Kelly, Pruitt and Su (2018), and Messmer (2017) further show the advantages and promises of…”
Dynamic factor model with infinite‐dimensional factor space: Forecasting (2016, methodology): “The Three-Pass Regression Filter, proposed in Kelly and Pruitt (2015), is based on the idea that, given the factors spanning the factor space of the dataset, it is possible to select those who are relevant in the prediction of a given target and discard those which are target-irrelevant.”
A deep learning based stock trading model with 2-D CNN trend detection (2017, neutral): “to perform time-series analysis [1], [3], [4].”