Gaussian Process Volatility Model
Authors: Yue Wu, José Miguel Hernández-Lobato, Zoubin Ghahramani
NeurIPS 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments with financial data show that GP-Vol performs significantly better than current standard alternatives. |
| Researcher Affiliation | Academia | Yue Wu Cambridge University wu5@post.harvard.edu Jose Miguel Hernandez Lobato Cambridge University jmh233@cam.ac.uk Zoubin Ghahramani Cambridge University zoubin@eng.cam.ac.uk |
| Pseudocode | Yes | Algorithm 1 RAPCF |
| Open Source Code | Yes | The code for RAPCF in GP-Vol is publicly available at http://jmhl.org. |
| Open Datasets | No | The paper uses "thirty daily Equity and twenty daily foreign exchange (FX) time series" but does not provide specific links, DOIs, or explicit statements of public availability with proper citations/sources for these datasets. |
| Dataset Splits | Yes | During the experiments, each method receives an initial time series of length 100. The different models are trained on that data and then a one-step forward prediction is made. The performance of each model is measured in terms of the predictive log-likelihood on the first return out of the training set. Then the training set is augmented with the new observation and the training and prediction steps are repeated. The whole process is repeated sequentially until no further data is received. |
| Hardware Specification | No | No specific hardware details (e.g., CPU/GPU models, memory) are provided for the experimental setup. |
| Software Dependencies | No | The paper mentions 'numerical optimization routines provided by Kevin Sheppard' but does not specify version numbers for any software dependencies. |
| Experiment Setup | Yes | We used GARCH(1,1), EGARCH(1,1) and GJR-GARCH(1,1,1) models... We placed broad non-informative priors on θ = (a, b, σn, γ, l) and used N = 200 particles and shrinkage parameter λ = .95 in RAPCF. |