# How to compute a confidence interval for the ratio of two population variances

## Description

Let’s say we want to compute a confidence interval for two population variances. We take two samples of data, $x_1, x_2, x_3, \ldots, x_k$ and $x’_1, x’_2, x’_3, \ldots, x’_k$, and compute their variances, $\sigma_1^2$ and $\sigma_2^2$. How do we compute a confidence interval for $\frac{\sigma_1^2}{\sigma_2^2}$?

Related tasks:

- How to compute a confidence interval for a mean difference (matched pairs)
- How to compute a confidence interval for a regression coefficient
- How to compute a confidence interval for a population mean
- How to compute a confidence interval for a single population variance
- How to compute a confidence interval for the difference between two means when both population variances are known
- How to compute a confidence interval for the difference between two means when population variances are unknown
- How to compute a confidence interval for the difference between two proportions
- How to compute a confidence interval for the expected value of a response variable
- How to compute a confidence interval for the population proportion

## Using SciPy, in Python

We’ll use R’s dataset EuStockMarkets as an example; of course you should replace this example data with your actual data when using this code. This dataset has information on the daily closing prices of 4 European stock indices. We’re going to compare the variability of Germany’s DAX and France’s CAC closing prices here. Let’s load in the dataset using the process explained in how to quickly load some sample data.

1
2
3
4
5
6
7
8
9
10

from rdatasets import data
import pandas as pd
# Load in the EuStockMarkets data and convert to a DataFrame
EuStockMarkets = data('EuStockMarkets')
df = pd.DataFrame(EuStockMarkets[['DAX', 'CAC']])
# Our two samples are its DAX and CAC columns
sample1 = df['DAX'].tolist()
sample2 = df['CAC'].tolist()

Now that we have our data loaded we can compute the confidence interval. You can change the confidence level by changing the value of $\alpha$ below.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18

# The degrees of freedom in each sample is its length minus 1
sample1_df = len(sample1) - 1
sample2_df = len(sample2) - 1
# Compute the ratio of the variances
import statistics
ratio = statistics.variance(sample1) / statistics.variance(sample2)
# Find the critical values from the F-distribution
from scipy import stats
alpha = 0.05 # replace with your chosen alpha (here, a 95% confidence level)
lower_critical_value = 1 / stats.f.ppf(q = 1 - alpha/2, dfn = sample1_df, dfd = sample2_df)
upper_critical_value = stats.f.ppf(q = 1 - alpha/2, dfn = sample2_df, dfd = sample1_df)
# Compute the confidence interval
lower_bound = ratio * lower_critical_value
upper_bound = ratio * upper_critical_value
lower_bound, upper_bound

1

(3.190589226470889, 3.827043522824141)

The 95% confidence interval for the ratio of the variances for Germany’s DAX and France’s CAC is $[3.191, 3.827]$.

Content last modified on 24 July 2023.

See a problem? Tell us or edit the source.

## Solution, in R

We’ll use R’s dataset EuStockMarkets as an example; of course you should replace this example data with your actual data when using this code. This dataset has information on the daily closing prices of 4 European stock indices. We’re going to compare the variability of Germany’s DAX and France’s CAC closing prices here.

1
2
3
4
5
6
7
8
9

# install.packages("datasets") # if you have not done so already
library(datasets)
# Load in the EuStockMarkets data and convert to a DataFrame
EuStockMarkets <- data.frame(EuStockMarkets)
# Our two samples are its DAX and CAC columns
sample.1 <- EuStockMarkets$DAX
sample.2 <- EuStockMarkets$CAC

Now that we have our data loaded we can compute the confidence interval. You can change the confidence level by changing the value of $\alpha$ below.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17

# The degrees of freedom in each sample is its length minus 1
df_1 = length(sample.1) - 1
df_2 = length(sample.2) - 1
# Compute the ratio of the variances
test.stat.ratio <- var(sample.1)/var(sample.2)
# Find the critical values from the F-distribution
alpha <- 0.05 # replace with your chosen alpha (here, a 95% confidence level)
lower_critical_value <- 1 / qf(p = alpha/2, df1 = df_1, df2 = df_2, lower.tail = FALSE)
upper_critical_value <- qf(p = alpha/2, df1 = df_2, df2 = df_1, lower.tail = FALSE)
# Compute the confidence interval and print it out
lower_bound <- test.stat.ratio*lower_critical_value
upper_bound <- test.stat.ratio*upper_critical_value
lower_bound
upper_bound

1
2
3
4
5

[1] 3.190589
[1] 3.827044

The 95% confidence interval for the ratio of the variances for Germany’s DAX and France’s CAC is $[3.191, 3.827]$.

Content last modified on 24 July 2023.

See a problem? Tell us or edit the source.

## Topics that include this task

## Opportunities

This website does not yet contain a solution for this task in any of the following software packages.

- Excel
- Julia

If you can contribute a solution using any of these pieces of software, see our Contributing page for how to help extend this website.