nosdata
Data package containing a collection of 'near-oracle' subgroups
Science Score: 44.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
○Academic publication links
-
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (8.9%) to scientific vocabulary
Repository
Data package containing a collection of 'near-oracle' subgroups
Basic Info
- Host: GitHub
- Owner: nickwkoning
- License: mit
- Language: R
- Default Branch: main
- Size: 52.7 MB
Statistics
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
- Releases: 0
Metadata Files
README.md
NOSdata
Data package containing a database of 'near-oracle' subgroups of the sign-flipping group. Each subgroup is represented as an n by m matrix, where n is the number of signs and m is the order of the group. Each column corresponds to an element of the group, under element-wise multiplication. The identity element is contained in the first column. The name of the near-oracle subgroup of order, say, m = 64 for n = 10 signs for one-sided and two-sided testing, respectively, is as follows.
{r}
max_abs_subgroups_n_10[[log2(64) + 1]]
max_subgroups_n_10[[log2(64) + 1]]
Construction
The package also contains two files that were used to generate the near-oracle subgroups (group_construction.R, groupconstructionparallel.R). These are based on the algorithm described in Koning & Hemerik (2022) implemented in the package NOS, and can be used to construct near-oracle subgroups that are not currently stored in the database.
Installation
You can install the data package using:
{r}
devtools::install_github("nickwkoning/NOSdata")
Example
Suppose we have n = 10 independent observations from a sign-flipping invariant distribution with mean .1 and standard deviation 1.
{r}
n = 10
mu = .1
x = mu + rnorm(n)
We want to test the null-hypothesis that the mean is zero against the alternative that it is non-zero at level .05. We use a sign-flipping subgroup of order 32 = 2^(6 - 1), which means it has rank 6. As we test against a two-sided alternative, we select a near-oracle subgroup whose name starts with "maxabs".
```{r}
alpha = .05
rank = 6
subgroup = get(paste0("maxabssubgroupsn", n))[[rank]]
To conduct the test, we compute the sign-flipped test statistics and compute the proportion of the test statistics on the sign-flipped data weakly exceed the test statistic on the actual data. The test statistic on the actual data coincides with the first element, as the identity is the first column of each subgroup. If this proportion is smaller than our level .05, we reject the null nypothesis.
{r}
teststats = t(subgroup) %*% x
pval = mean(teststats[1] <= teststats)
pval < alpha
In case we would like to conduct a one-sided test, we use the following subgroup.
{r}
subgrouponesided = get(paste0("maxsubgroupsn_", n))[[rank]]
```
Owner
- Name: Nick
- Login: nickwkoning
- Kind: user
- Repositories: 1
- Profile: https://github.com/nickwkoning
Citation (CITATION.cff)
cff-version: 1.2.0
message: "If you use this software, please cite it as below."
authors:
- family-names: Koning
given-names: N.W.
title: "Database of Near Oracle Subgroups"
version: 0.1
date-released: 2022-01-25
url: "https://github.com/nickwkoning/NOSdata"
GitHub Events
Total
Last Year
Dependencies
- R >= 2.10 depends