Concept Architecture
Concept
Theoretically, F-Distribution is a continuous probability distribution that describes the ratio of two independent chi-square random variables, each divided by its respective degrees of freedom. It is derived from sampling theory and forms the basis of many inferential statistical procedures that compare variances or evaluate overall model fit. In health economics, the F-distribution underpins analysis of variance, regression modelling and numerous hypothesis tests used in clinical and economic evaluations.
Mathematically, the F-distribution is defined as the ratio of two scaled independent chi-square distributions. Its shape depends on two parameters corresponding to the numerator and denominator degrees of freedom. Critical values and cumulative probabilities from the distribution are used to determine statistical significance for variance comparisons and model evaluation.
In practice, the F-distribution is applied by calculating an F-statistic from observed data and comparing it with the theoretical F-distribution having the appropriate degrees of freedom. It is routinely used in health economics when comparing group means, testing overall regression significance and assessing nested statistical models.
Purpose
Used to determine the statistical significance of variance ratios, compare group means, evaluate regression models and support hypothesis testing in health economic analyses.
Mathematical Formulae
Primary Formula
F = (??� / ??) � (??� / ??)
where:
- ??� = first independent chi-square random variable
- ??� = second independent chi-square random variable
- ?? = numerator degrees of freedom
- ?? = denominator degrees of freedom
Supporting Formulae
For analysis of variance:
F = MS?between? � MS?within?
MS = SS � df
where:
- MS = mean square
- SS = sum of squares
- df = degrees of freedom
Related Mathematical Methods
- Analysis of Variance (ANOVA)
- Linear Regression
- Nested Model Comparison
- Variance Components Analysis
- General Linear Model
- Hypothesis Testing
Example
A health economist compares mean annual healthcare costs across four treatment groups. The between-group mean square is �2,400 and the within-group mean square is �800.
F = 2,400 � 800 = 3.00
This F-statistic is compared with the F-distribution using the appropriate numerator and denominator degrees of freedom. If the calculated value exceeds the critical value at the chosen significance level, the null hypothesis that all group means are equal is rejected.
Excel Implementation
| Function | Example Formula | Health Economics Application |
|---|---|---|
| F.DIST | =F.DIST(A2,3,96,TRUE) | Calculate cumulative F probability |
| F.DIST.RT | =F.DIST.RT(A2,3,96) | Calculate right-tail probability for an F-statistic |
| F.INV | =F.INV(0.95,3,96) | Obtain F critical value |
| F.INV.RT | =F.INV.RT(0.05,3,96) | Determine upper-tail critical value for hypothesis testing |
VBA (Optional)
Automate ANOVA and regression analyses by calculating F-statistics, determining critical values and producing formatted hypothesis test reports.
Sources
- Casella G, Berger RL. Statistical Inference.
- Rice JA. Mathematical Statistics and Data Analysis.
- Kutner MH, Nachtsheim CJ, Neter J, Li W. Applied Linear Statistical Models.
- Draper NR, Smith H. Applied Regression Analysis.
- Drummond MF, et al. Methods for the Economic Evaluation of Health Care Programmes.
Related Concepts (2)
Library
Publications
1
Statistical Analysis of Cost-Effectiveness Data — Willan & Briggs, 1st Edition ed., 2006 (John Wiley & Sons)
A synthesis of statistical methods for analysing cost-effectiveness data, including net-benefit regression, confidence intervals for the ICER, cost-effectiveness acceptability curves, and covariate adjustment. Part of the Wiley Statistics in Practice series.
BookView source →
Frequently Asked Questions (6)
What is the F-distribution?
A continuous probability distribution arising from the ratio of two independent chi-square variables divided by their degrees of freedom, underlying analysis of variance.
Source: Fisher 1925
What does the F-distribution allow researchers to test?
The F-distribution arises from the ratio of two independent chi-square quantities, each divided by its degrees of freedom, and it provides the reference for tests that compare variances. It allows researchers to test whether the variation between several group means is large relative to the variation within groups, which is the core of analysis of variance. An observed ratio is compared with this distribution to judge whether the group differences exceed what chance would produce. Supplying the yardstick for variance-ratio tests is its role. Kirkwood and Sterne (2003) describe this distribution.
Source: Kirkwood & Sterne 2003
How is the F-distribution used?
The F-distribution is used as the reference distribution for tests whose statistics are ratios of variances or of mean squares, most notably the analysis of variance, which compares the variation between groups with that within groups, and the F-test for comparing two variances or the overall significance of a regression. So the F-distribution is used to assess whether variability attributable to one source significantly exceeds another, by referring the computed F-statistic to the distribution with the appropriate degrees of freedom, which is why it is fundamental to analysis of variance, regression, and other procedures that compare sources of variation.
Source: Fisher 1925
What are the degrees of freedom in the F-distribution?
The F-distribution has two degrees-of-freedom parameters, one for the numerator and one for the denominator of the ratio, corresponding to the two chi-square variables that form it. These determine the exact shape of the distribution. So the degrees of freedom of the F-distribution come in a pair, reflecting the numerator and denominator, and in tests they derive from the structure of the data, such as the number of groups and the sample size in analysis of variance, which is why identifying both degrees of freedom correctly is necessary to refer an F-statistic to the right F-distribution and obtain a valid p-value.
Source: Fisher 1925
How does the F-distribution relate to analysis of variance?
The F-distribution relates to analysis of variance as the distribution of its test statistic: analysis of variance compares the mean square between groups with the mean square within groups by forming their ratio, the F-statistic, which follows the F-distribution under the null hypothesis of no difference between group means. So the F-distribution provides the reference against which the F-statistic from analysis of variance is judged, allowing a p-value to be obtained for whether the between-group variation significantly exceeds the within-group variation, which is why the F-distribution is central to testing differences among group means in analysis of variance.
Source: Fisher 1925
How does the F-distribution relate to the chi-square distribution?
The F-distribution relates to the chi-square distribution because it is formed from the ratio of two independent chi-square variables, each divided by its degrees of freedom. Thus the F-distribution is built directly from chi-square components, and as the denominator degrees of freedom grow large, the F-distribution approaches a scaled chi-square. So the F-distribution and the chi-square distribution are closely connected, with the F arising as a ratio of scaled chi-squares, which links the distributions used for comparing variances to those used for goodness of fit and other squared-deviation tests, all part of the family of distributions derived from the normal.
Source: Fisher 1925
Trust Record
Verified by Dr Darrin Baines
British health economist
Professional identity: darrinbaines.org
Verification date: 16 Dec 2025
Content version: 1.0.0
Canonical Identity
- Persistent URI
- https://healtheconomics.wiki/concept/f-distribution
- Term code
- HE-ES-SA-064
Stable URI · Machine-readable · Resolvable · CC BY 4.0