Stack Overflow
Stack Overflow

Questions tagged [quantile-regression]

107 questions
10
votes
2 answers

Quantile random forests from scikit-garden very slow at making predictions

I've started working with quantile random forests (QRFs) from the scikit-garden package. Previously I was creating regular random forests using RandomForestRegresser from sklearn.ensemble. It appears that the speed of the QRF is comparable to the…
Tim Williams
  • 183
  • 1
  • 10
5
votes
0 answers

Quantile Regression generates non-monotonic quantile forecasts, e.g. Q49>Q50

I would expect that a quantile regression gives forecasts for the quantiles that are monotonic, i.e. However, the quantreg package in R generates forecasts that do not make sense at all, see the plot: Is there a reason for that? Example below.…
HOSS_JFL
  • 765
  • 2
  • 9
  • 24
5
votes
1 answer

QuantileRegression ValueError: operands could not be broadcast together with shapes

I am trying to forecast my target variable using Quantile Regression in Python. The data I am considering for training and validation is from period 2015 Oct -2017 Dec 31st. Now the model has developed,Im trying to forecast values for 2018 Jan,…
Nithin Das
  • 364
  • 1
  • 5
  • 14
5
votes
0 answers

Difference between rqPen and quantreg packages in R

I am building a quantile regression model + LASSO penalization for the boston housing data in R. I found 2 packages that can build this kind of models: rqPen and quantreg. rqPen implements a cross validation process to tune the LASSO parameter…
Álvaro Méndez Civieta
  • 725
  • 3
  • 12
  • 33
5
votes
1 answer

Fast nonnegative quantile and Huber regression in R

I am looking for a fast way to do nonnegative quantile and Huber regression in R (i.e. with the constraint that all coefficients are >0). I tried using the CVXR package for quantile & Huber regression and the quantreg package for quantile…
Tom Wenseleers
  • 7,535
  • 7
  • 63
  • 103
4
votes
2 answers

Is it allowed/possible to call an R function or fortran code within a pragma openmp parallel for loop in Rcpp?

In an Rcpp project I would like to be able to either call an R function (the cobs function from the cobs package to do a concave spline fit) or call the fortran code that it relies on (the cobs function uses quantreg's rq.fit.sfnc function to fit…
Tom Wenseleers
  • 7,535
  • 7
  • 63
  • 103
3
votes
1 answer

How can I extend the quantile regression lines geom_quantile to forecast in ggplot?

I am trying to plot the quantile regression lines for a set of data. I would like to extend the quantile regression lines from geom_quantile() in order to show how they forecast similar to using stat_smooth() with the fullrange argument set to TRUE.…
joerminer
  • 153
  • 8
3
votes
2 answers

Why its takes so much longer to fit model in sklearn.linear_model.QuantileRegressor then R model implementation?

First I used R implementation quantile regression, and after that I used Sklearn implementation with the same quantile (tau) and alpha=0.0 (regularization constant). I am getting the same formulas! I tried many "solvers" and still the running time…
Sapir Tubul
  • 31
  • 3
3
votes
1 answer

summary.rq output depends on sample size

I found out (see below) that the function summary.rq (page 88) from the quantreg package prints different output, depending on whether sample size is greater equal or less than 1001. I am aware, that rq() uses a different method depending on whether…
Buochserhorn
  • 133
  • 4
3
votes
1 answer

Identifying Outliers with Quantile Regression and Python

I am trying to identify outliers in a dataset using the 5th and 95th percentiles of a regression line so I'm using quantile regression in Python with statsmodel, matplotlib and pandas. Based on this answer from blokeley, I can create a scatterplot…
Matt
  • 203
  • 1
  • 4
  • 8
3
votes
1 answer

R - Setting margins does not work

I estimated a linear regression model by using the quantreg package. I now want to display the results graphically by using the plot() function: plottest = plot(summary(QReg_final), parm=2, main="y",ylab="jjjjj") The result is the following (I…
shenflow
  • 345
  • 2
  • 12
2
votes
1 answer

How to model quantiles regression curves for probabilities depending on a predictor in R?

I'd' like to model the 25th, 50th and 75th quantile regression curves (q25, q50, q75) for 241 values of probability ('prob') depending on x0. For that purpose, I used the qgamV package as follows. However, this approach led to some q25, q50, q75…
denis
  • 199
  • 1
  • 8
2
votes
1 answer

Confusion about the matrix "B" returned by `quantreg::boot.rq`

When invoking boot.rq like this b_10 = boot.rq(x, y, tau = .1, bsmethod = "xy", cov = TRUE, R = reps, mofn = mofn) what does the B matrix (size R x p) in b_10 contain: bootstrapped coefficient estimates or bootstrapped standard errors? The Value…
erised
  • 187
  • 7
2
votes
1 answer

Is it possible to use lqmm with a mira object?

I am using the package lqmm, to run a linear quantile mixed model on an imputed object of class mira from the package mice. I tried to make a reproducible…
De La Cruz
  • 33
  • 4
2
votes
3 answers

Is there a neat approach to label a ggplot plot with the equation and other statistics from geom_quantile()?

I'd like to include the relevant statistics from a geom_quantile() fitted line in a similar way to how I would for a geom_smooth(method="lm") fitted linear regression (where I've previously used ggpmisc which is awesome). For example, this code: #…
Mark Neal
  • 996
  • 16
  • 52
1
2 3 4 5 6 7 8