Variable Selection in Predictive Regressions

By Serena Ng. 2013

66 citations recorded by the source corpus.

Subfields: Applied Mathematics, Economics and Econometrics, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/1321b86c5d5a41fe76302f8d8b711ffc1ae26747

How researchers see this paper

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

In Applied Mathematics

1. Yongli Zhang
2. Shinya Tanaka
3. Xun Lu
4. Xu Cheng
5. Liangjun Su
6. Yoshimasa Uematsu
7. Yuhong Yang
8. B. Hansen

In Economics and Econometrics

1. Liangjun Su
2. Xun Lu
3. Xu Cheng
4. Yongli Zhang
5. Shinya Tanaka
6. Yoshimasa Uematsu
7. Yuhong Yang
8. B. Hansen

In Statistics and Probability

1. Xun Lu
2. Liangjun Su
3. Yongli Zhang
4. Xu Cheng
5. Xinyu Zhang
6. B. Hansen
7. Guohua Zou
8. Yuhong Yang

Citations in the corpus

• Jackknife model averaging for quantile regressions (2015, background): “As Ng (2013) puts it in her abstract, “(i)rrespective of the model size, there is an unavoidable tension between prediction accuracy and consistent model determination.””
• High‐dimensional macroeconomic forecasting and variable selection via penalized regression (2019, background): “In terms of variable selection in predictive regressions, Ng (2013) investigates finite sample properties of forecast values based on the BIC, AIC and Lasso, and finds the Lasso has rather stable compared to the others.”
• Cross-validation for selecting a model selection procedure (2015, background): “See, e.g., Fan et al (2011) and Ng (2013) for recent reviews and discussions of model selection methods in the traditional and high-dimensional settings for model identification and prediction.”
• Forecasting with Factor-Augmented Regression: A Frequentist Model Averaging Approach, Second Version (2013, background): “This is in contrast to inference on the coeffi cients, where Pagan (1984), Bai and Ng (2009), Ludvigson and Ng (2011), and Gonçalves and Perron(2011) have shown that the generated regressors affect the sampling distribution.”
• On LASSO for predictive regression (2018, unknown): “2 Performance We apply the set of feasible forecasting methods as in Section 4 to forecast short-horizon and longhorizon stock returns recursively with either a 10-year or 15-year rolling window. All 12 variables are made available in the predictive regression, to which Welch and Goyal (2008) refer ”