A Quantile Regression Neural Network Approach to Estimating the Conditional Density of Multiperiod Returns

By James W. Taylor. 2000

389 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

1. Dhruv Madeka
2. David Salinas
3. Balakrishnan Narayanaswamy
4. Matteo Sesia
5. Kari Torkkola
6. E.J. Candes
7. Ruofeng Wen
8. Jan Gasthaus

In Finance

1. David Salinas
2. Matteo Sesia
3. E.J. Candes
4. Valentin Flunkert
5. Jan Gasthaus
6. Alex J. Cannon
7. Yaniv Romano
8. Stephan Rasp

In Management Science and Operations Research

1. Bernie Wang
2. Tim Januschowski
3. Jan Gasthaus
4. Syama Sundar Rangapuram
5. Dhruv Madeka
6. David Salinas
7. Bryan Lim
8. Balakrishnan Narayanaswamy

Citations in the corpus

• A Multi-Horizon Quantile Recurrent Forecaster (2017, background): “QR is robust since it does not make distributional assumptions, produces accurate probabilistic forecasts with sharp prediction intervals, and often serves as a post-processor for prediction calibration (Taylor, 2000).”
• Conformalized Quantile Regression (2019, methodology): “In our experience, quantile regression forests [22] are often overly conservative and quantile neural networks [20] are occasionally so.”
• With Malice Towards None: Assessing Uncertainty via Equalized Coverage (2019, neutral): “Then, consider any algorithm A for quantile regression that estimates conditional quantile functions from observational data, such as quantile neural networks [31] (described in Appendix A).”
• Probabilistic Forecasting with Spline Quantile Function RNNs (2019, background): “One of the first approaches that uses neural networks to estimate quantiles appears in (Taylor, 2000) where a quantile regression neural network (QRNN) is proposed.”
• Quantile regression neural networks: Implementation in R and application to precipitation downscaling (2011, background): “The result is the QRNN model (Taylor, 2000), which is a flexible model that can represent nonlinear predictor–predictand relationships, includingones involving interactions between predictors, without prior specification of the form of the relationships by the modeller.”
• A comparison of some conformal quantile regression methods (2019, methodology): “Any algorithm can be employed for this purpose; for example, one may rely on linear regression(Koenker & Bassett Jr, 1978), neural networks(Taylor, 2000), or random forests(Meinshausen, 2006).”
• Neural networks for post-processing ensemble weather forecasts (2018, methodology): “Other probabilistic neural network approaches include quantile regression (Taylor, 2000) and distribution-to-distribution regression (Kou et al., 2018).”