Derivative Estimation in Random Design
Authors: Yu Liu, Kris De Brabanter
NeurIPS 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In the simulation study, we show that the new estimator has similar performance compared to local polynomial regression and penalized smoothing splines. |
| Researcher Affiliation | Academia | Yu Liu1, Kris De Brabanter1,2 1Department of Computer Science, 2Department of Statistics Iowa State University, Ames, IA 50011, USA. |
| Pseudocode | No | The paper does not contain any pseudocode or algorithm blocks. |
| Open Source Code | No | The paper cites external tools (ks, locpol, pspline) with their URLs, but there is no explicit statement from the authors about releasing the source code for their own described methodology. |
| Open Datasets | No | The paper uses simulated data generated according to mathematical functions (e.g., 'm(X) = cos2(2πX) for X beta(2, 2)', 'X(1 X) sin{(2.1π)/(X + 0.05)} for X U(0.25, 1)'). It does not use or provide access to a public, external dataset. |
| Dataset Splits | No | The paper constructs data sets for Monte Carlo simulations ('constructed data sets of size n = 700 and generated the functions... 100 times'), but it does not specify explicit train/validation/test splits for these generated datasets or any cross-validation setup. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware used for running the experiments (e.g., CPU, GPU models, or memory). |
| Software Dependencies | Yes | The paper mentions specific software with version numbers: 'ks: Kernel smoothing v1.11.1', 'locpol: Kernel local polynomial regression v0.6', 'pspline: Penalized smoothing splines v1.0-18'. |
| Experiment Setup | Yes | The tuning parameter k is selected over the integer set [1, (n - 1)/2 ] and according to Corollary 2. We use local cubic regression (p = 3) with bimodal kernel to initially smooth the data. Bandwidths for the bimodal kernel ˆhb are selected from the set {0.1, 0.105, 0.11, . . . , 0.2} and corrected for a unimodal Gaussian kernel. Bandwidths are selected from the set {0.04, 0.045, . . . , 0.08} and corrected for a unimodal Gaussian kernel. The order of the local polynomial is set to p = 2... sample size n = 700 and e N(0, 0.22)... constructed data sets of size n = 700 and generated the functions... 100 times according to model (2) with e N(0, 0.22) and e N(0, 0.32) for model (12) and model (13) respectively. |