# qrnn v2.0.5

0

0th

Percentile

## Quantile Regression Neural Network

Fit quantile regression neural network models with optional left censoring, partial monotonicity constraints, generalized additive model constraints, and the ability to fit multiple non-crossing quantile functions following Cannon (2011) <doi:10.1016/j.cageo.2010.07.005> and Cannon (2018) <doi:10.1007/s00477-018-1573-6>.

## Functions in qrnn

 Name Description dummy.code Convert a factor to a matrix of dummy codes gam.style Modified generalized additive model plots for interpreting QRNN models qrnn2 Fit and make predictions from QRNN models with two hidden layers qrnn.initialize Initialize a QRNN weight vector transfer Transfer functions and their derivatives tilted.abs Tilted absolute value function quantile.dtn Interpolated quantile distribution with exponential tails YVRprecip Daily precipitation data at Vancouver Int'l Airport (YVR) censored.mean A hybrid mean/median function for left censored variables composite.stack Reformat data matrices for composite quantile regression adam Adaptive stochastic gradient descent optimization algorithm (Adam) qrnn.predict Evaluate quantiles from trained QRNN model qrnn.cost Smooth approximation to the tilted absolute value cost function qrnn.fit Main function used to fit a QRNN model or ensemble of QRNN models qrnn-package Quantile Regression Neural Network qrnn.rbf Radial basis function kernel huber Huber norm and Huber approximations to the ramp and tilted absolute value functions mcqrnn Monotone composite quantile regression neural network (MCQRNN) for simultaneous estimation of multiple non-crossing quantiles No Results!