Semiparametric Inference Using Fractional Posteriors
Authors: Alice L'Huillier, Luke Travis, Ismaël Castillo, Kolyan Ray
JMLR 2023 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We also perform simulations which confirm that the derived asymptotic theoretical properties are empirically relevant and observable at reasonable finite sample sizes: in particular, we illustrate that the modified credible sets have close to optimal coverage already at moderate sample size. ... Simulation study. We now illustrate the applicability of the asymptotic result presented in Proposition 3.2 to the finite sample setting. We simulated 10,000 observations of Y n from the Gaussian white noise model (n = 10, 000) with 3 different parameter combinations of (β, µ, γ) corresponding to the three different cases presented in Proposition 3.2. ... This data is presented in Table 1. |
| Researcher Affiliation | Academia | Alice L Huillier EMAIL LPSM, Sorbonne Universit e 4, place Jussieu 75005, Paris, France Luke Travis EMAIL Department of Mathematics Imperial College London London SW7 2AZ, United Kingdom Isma el Castillo EMAIL LPSM, Sorbonne Universit e 4, place Jussieu 75005, Paris, France Kolyan Ray EMAIL Department of Mathematics Imperial College London London SW7 2AZ, United Kingdom |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. It describes mathematical methods and proofs in prose and equations. |
| Open Source Code | No | Posterior draws were generated by MCMC using the sbde package (Tokdar et al., 2022). ... We again generate posterior samples by MCMC using the sbde R-package (Tokdar et al., 2022). The paper mentions using a third-party R-package ('sbde') but does not provide any information about releasing its own source code for the methodology described. |
| Open Datasets | No | We simulated 10,000 observations of Y n from the Gaussian white noise model (n = 10, 000)... Consider density estimation with n = 10, 000 observations drawn from the density on [0, 1] given by f eg... The paper uses simulated data based on theoretical models (Gaussian white noise, density estimation) rather than publicly available datasets. |
| Dataset Splits | No | The paper simulates data rather than using a pre-existing dataset. It mentions generating '10,000 observations' but does not specify any training/test/validation splits for this simulated data. |
| Hardware Specification | No | The paper mentions performing simulations but does not provide any specific details about the hardware used (e.g., GPU/CPU models, memory specifications). |
| Software Dependencies | No | Posterior draws were generated by MCMC using the sbde package (Tokdar et al., 2022). The paper mentions the 'sbde package' but does not specify its version number or any other software dependencies with version information. |
| Experiment Setup | Yes | We simulated 10,000 observations of Y n from the Gaussian white noise model (n = 10, 000) with 3 different parameter combinations of (β, µ, γ) corresponding to the three different cases presented in Proposition 3.2. ... Figure 1 displays histograms of αn posterior draws of nαn(ψ(f) ˆψ)/ ψf0 L with αn = 1/4 for combinations of β, γ; ... Figure 2 shows the roughly Gaussian shape of each of the posterior distributions; ... with αn = n 1/4/ p log(n), n 1/4p log(n) and 1. |