XNomial: Exact Goodness-of-Fit Test for Multinomial Data

Bill Engels authored this statistical package for testing whether observed counts fit a given expected ratio. The description suggests it is used for applications like analyzing genetic cross data against expected frequencies. It provides exact tests via enumeration or Monte Carlo sampling, using statistics like the log-likelihood ratio or chi-square.

Use Cases

• Test genetic cross outcomes against expected Mendelian ratios based on the example in the description
• Perform exact goodness-of-fit analysis for multinomial count data using the xmulti() function
• Conduct Monte Carlo sampling for large multinomial tests using the xmonte() function
• Compare deviation measures like log-likelihood ratio or chi-square statistic as described

Strengths

• Provides exact tests for multinomial goodness-of-fit, which can be more accurate than asymptotic approximations
• Offers multiple measures of deviation including log-likelihood ratio, multinomial probability, and chi-square statistic
• Includes functionality for both full enumeration and random sampling of outcomes

Limitations

• Row count, column definitions, and sample data are unknown, limiting suitability assessment
• Description metadata is limited; actual data quality requires manual inspection after download
• Last update date is unknown; freshness unverified

Provenance

Source

Bill Engels