Questions tagged [quantile-regression]
Quantile regression allows us to estimate the effect of a set of predictor variables over the entire distribution of the outcome variable or any particular quantile.
418 questions
1
vote
0
answers
26
views
Does compositional structure (actually) mitigate the curse of dimensionality?
The paper "Deep Quantile Regression: Mitigating the Curse of Dimensionality Through Composition" makes the following claim (top of page 4):
It is clear that smoothness is not the right ...
- 322
0
votes
0
answers
36
views
Conditional quantile regression versus "conditional" conformal prediction
I ask this question with the comments of Section 2.3 of "Conformal Prediction with Conditional Guarantees" in mind. I'm not fully familiar with non-parametric methods for quantile regression ...
- 322
4
votes
1
answer
303
views
Quantile gam (qgam) of tensor product using gratia: drawing a contour plot of the response rather than the partial effect
Let's imagine the quantile regression (qgam) of the tensor product below.
...
- 295
0
votes
0
answers
107
views
Systematic underperformance for higher quantiles - How to fix
TLDR; Systematic underperformance for higher quantiles (Q50, Q75, Q90), both in terms of relative pinball loss and PIT density. How can I turn this around?
Question
Why do my probabilistic forecasts ...
- 1
0
votes
0
answers
75
views
how to determine the R value in linear quantile mixed modeling using R?
We are doing a linear quantile mixed modeling using R, if our understanding is correct, R in the lqmm package means the number of bootstrap replications. In our model, we have 4 Level 2 predictors and ...
-1
votes
2
answers
191
views
Quantile regression: But which quantiles (conditional on what)?
I have a task of making a quantile regression (5%, 50% and 95%) for tomorrow's power production. However, I am trying to grasp which quantiles we are talking about. Wikipedia (and similar sites) ...
0
votes
0
answers
31
views
Estimating Median Treatment Effect on matched data from MatchIt package in R?
I've been using the MatchIt package in R to estimate average treatment effect on treated (ATT) for health insurance programs with observational data.
Can I get ...
- 6,644
4
votes
1
answer
135
views
Is the quantile regression predicting the unconditional quantile at $\bar x$?
We know that a mathematical property of the OLS estimator is that predicts the mean of Y at the mean of X. For example, in model $y = \beta_0 + \beta_1 x + \mu$ we have that:
$$\bar{y} = \hat{\beta_0^{...
- 2,830
4
votes
1
answer
106
views
Applying quantile regression to naturally log-transform dependent variable and interpret results as symmetrical percentage change
I am using quantile regression and I was wondering whether it is appropriate to apply a natural log transformation of the dependent variable and then interpret the quantile regression results as ...
- 91
1
vote
2
answers
90
views
Quantile loss function in lightGBM: unexpected values for prediction intervals
I am trying to fit a lightGBM regression model to my data, using quantile as the loss function, but the results are strange.
In particular, I want to estimate the median value of my response variable ...
- 328
0
votes
0
answers
110
views
Minimum number of observations quantile regression
My goal is to estimate the market beta (so exposure of an asset returns to market shocks) in quantiles :
$Q_{r_i|r_M} = a_0(\tau) + \beta_i(\tau)r_M+\varepsilon_i(\tau)$
where $r_i$ are asset returns (...
- 130
0
votes
0
answers
74
views
Effect size for quantile regression
How should effect size be computed for a quantile regression?
The formula for Cohen's d depends on pooled standard deviation, which depends on sample sizes (both explicitly and via the separate sample ...
- 101
5
votes
1
answer
232
views
Quantile Regression Coefficients Equality Across Quantiles
I know that we can test the coefficient equality across different quantiles $\tau$ in R with an anova :
...
- 130
1
vote
0
answers
56
views
Generating a conditional density function of a response variable from regression quantiles
In R, I'm using the quantreg package to generate regression quantiles to examine the full conditional distribution of a response variable. The scatter plot is wedge shape (non-negative with ...
- 11
5
votes
0
answers
158
views
Rate of convergence of $\ell_1$-penalized quantile regression is $\sqrt{\frac{s\log (p \vee n)}{n}}$
In the standard LASSO literature, you often encounter that the LASSO estimator converges at a rate of $\sqrt{\frac{s\log p}{n}}$ (see e.g. this post).
A related method is the $\ell_1$-penalized ...
- 724