Talk:Quantile regression

Reorder point (inventory) as quantile estimation of the demand

The article does not discuss much about the practical application of quantiles. For retail and manufacturing, quantile regression is very handy for inventory management because it basically represents the reorder point in an inventory optimization context, see http://www.lokad.com/reorder-point-definition.ashx Disclaimer: I am a shareholder of Lokad, and might have a biased POV (so I prefer to let someone else deal with this part of the discussion on the primary article). --Joannes Vermorel (talk) 15:23, 23 February 2012 (UTC)[reply]

In the intuition section, I think the amount of change in the expected loss due to the shift of q by 1 is wrong. I get 0.5 times the integral of the density from negative infinity to q, minus 0.5 times the integral of the density from q to infinity, plus the integral of y-q-1 times the density from q to q+1. Even if what is written on the article now is correct, some explanation would be appreciated. Tet21tet (talk) 06:03, 27 May 2014 (UTC)[reply]

In the Quantile section, in the 4th line after "Define the loss function as...", the indicator function appears incorrect. Shouldn't it be I{y<u}? 2620:0:1000:167C:5891:BE8:3B61:E2B (talk) 16:07, 10 June 2015 (UTC)[reply]

Definitions

Request: Define Equivariance in the Wikipedia article. — Preceding unsigned comment added by 64.128.27.82 (talk) 19:43, 16 April 2012 (UTC)[reply]

Dr. Allen's comment on this article

Dr. Allen has reviewed this Wikipedia page, and provided us with the following comments to improve its quality:

I think the piece could emphasize that the method involves taking the absolute value of deviations around the median as opposed to squared deviations around the mean. This helps make it less sensitive to outliers. Some reference should be made to its applications in risk analysis, especially given the Basel Accords adoption of Value at Risk (VaR) which is based on breaches of a given quantile. Nobel laurate Robert Engle and Simone Manganelli have their CAViaR model (2004) "CAViaR: Conditional Autoregressive Value at
Risk by Regression Quantiles", Journal of Business and Economic Statistics, 22,4, pp: 367-381. Some reference could also be made to the use of quantile regression in fitting copulae. See for example: Bouyé, E., & Salmon, M. (2009). Dynamic copula quantile regressions and tail area dynamic dependence in Forex markets. The European Journal of Finance, 15 (7-8),
721-750.

We hope Wikipedians on this talk page can take advantage of these comments and improve the quality of the article accordingly.

Dr. Allen has published scholarly research which seems to be relevant to this Wikipedia article:

Reference : D.E. Allen & Abhay K Singh & R. Powell & Michael McAleer & James Taylor & Lyn Thomas, 2012. "The Volatility-Return Relationship: Insights from Linear and Non-Linear Quantile Regressions," Documentos de Trabajo del ICAE 2012-24, Universidad Complutense de Madrid, Facultad de Ciencias Economicas y Empresariales, Instituto Complutense de Analisis Economico.

ExpertIdeasBot (talk) 06:11, 9 July 2015 (UTC)[reply]

Inclusion of random effects

There is currently no information on including random effects in quantile regression models in this article. Marco Geraci has an R package lqmm (and recent add ons aqmm and nlqmm) for linear (nonlinear and additive) mixed-effects quantile regression models. It would be wonderful if these and the thinking behind them had some coverage on this page ^[1]. Muniche (talk) 15:04, 14 August 2018 (UTC)[reply]

References

^ Geraci, Marco; Bottai, Matteo (20 February 2013). "Linear quantile mixed models". Statistics and Computing. 24 (3): 461–479. doi:https://doi.org/10.1007/s11222-013-9381-9. {{cite journal}}: Check |doi= value (help); External link in |doi= (help)

Proposed merge of Repeated median regression into Quantile regression

as per lead: "quantile regression estimates the conditional median..." fgnievinski (talk) 20:11, 26 August 2022 (UTC)[reply]

They're not the same. 0.5-quantile regression is a method for constructing a conditional median line of best fit over points

(X_{i},y_{i})

that reduces to the standard median when

X=1

. Repeated median regression is a method for constructing a robust line of best fit over points

(X_{i},y_{i})

that uses nested medians. They have different properties, e.g. 0.5-quantile regression has a breakdown point of 0, whereas repeated median regression has a breakdown point of 0.5. Preimage (talk) 03:57, 9 October 2022 (UTC)[reply]

Closing, with no merge, given the uncontested objection and no support. Klbrain (talk) 08:33, 23 November 2022 (UTC)[reply]

[1] Geraci, Marco; Bottai, Matteo (20 February 2013). "Linear quantile mixed models". Statistics and Computing. 24 (3): 461–479. doi:https://doi.org/10.1007/s11222-013-9381-9. {{cite journal}}: Check |doi= value (help); External link in |doi= (help)

[1]