Statistical Insights into HSIC in High Dimensions
Authors: Tao Zhang, Yaowu Zhang, Tingyou Zhou
NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We also conduct extensive numerical studies to validate our theoretical results. ... In this subsection, we conduct some simulation studies to validate our theoretical conclusions on the HSIC based test in high dimensions. ... The empirical powers of different tests for Models (I)-(III) in Example 2. ... The empirical powers of different tests for Models (IV)-(VI) in Example 3. ... We are interested in whether there is any dependency between stock prices of the energy sector and the raw material sector in the U.S. stock market. |
| Researcher Affiliation | Academia | Tao Zhang School of Information Management and Engineering Shanghai University of Finance and Economics Shanghai 200433, China ... Yaowu Zhang School of Information Management and Engineering Mo E Key Laboratory of Interdisciplinary Research of Computation and Economics Shanghai University of Finance and Economics Shanghai 200433, China ... Tingyou Zhou School of Data Sciences Zhejiang University of Finance and Economics Hangzhou 310018, China |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any concrete access to source code for the methodology described. |
| Open Datasets | Yes | To see this, we extract the monthly mean stock prices of energy companies as well as raw material companies starting from January 2021 to December 2022 from https://finance.yahoo.com/. |
| Dataset Splits | No | The paper uses generated synthetic data for simulations and a collected real-world dataset, but does not provide specific training/validation/test dataset split information (percentages, sample counts, or citations to predefined splits) needed to reproduce the data partitioning. |
| Hardware Specification | No | The paper does not provide any specific hardware details (exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) used for running its experiments. |
| Software Dependencies | No | The paper mentions using specific kernels and implementing tests but does not provide any specific ancillary software details (e.g., library or solver names with version numbers like Python 3.8, CPLEX 12.4) needed to replicate the experiment. |
| Experiment Setup | Yes | Throughout the simulations, we choose two kinds of commonly used kernels to implement the HSIC based tests, i.e., Gaussian and Laplacian. ... For both choices of kernels, we set the bandwidth parameters to be γx = c0γm x and γy = c0γm y and vary c0 from 0.5 to 2. Here, γm z represents the median of { zi zj }1 i<j n, and z is either x or y. Due to space constraints, we present results specifically for the case when γx = γm x and γy = γm y . ... We fix the sample size to be 100 and consider two scenarios for the covariate dimensions. |