One-sample z-test for means

Use case

We want to estimate the population mean with regard to a single, quantitave variable. We know with certainty the population standard deviation, \(\sigma\). We have a sample of size \(n\) and popluation mean \(\bar{x}\).

Preconditions

To estimate the population mean using the normal distribution, one of the following conditions must hold:

  1. The population is normally distributed.
  2. The sample size, \(n\), is large enough for the Central Limit Theorem to take effect.

Hypotheses

Null hypothesis, \(H_0\): \(\mu = \mu_0\)

Alternative hypothesis, \(H_a\): \(\mu \ne/>/< \mu_0\)

Test statistic

\[ z = \frac{\bar{x} - \mu_0}{\frac{\sigma}{\sqrt{n}}} \]

p-value

Where Z is a standard normal random variable,

  1. For \(H_a : \mu < \mu_0\): \(p = P(Z \le z)\)
  2. For \(H_a : \mu > \mu_0\): \(p = P(Z \ge z)\)
  3. For \(H_a : \mu \ne \mu_0\): \(p = 2P(Z \le -|z|)\)

p-value (Python)

  1. For \(H_a : \mu < \mu_0\): p = stats.norm.cdf(z)
  2. For \(H_a : \mu > \mu_0\): p = 1 - stats.norm.cdf(z)
  3. For \(H_a : \mu \ne \mu_0\): p = 2 * stats.norm.cdf(-abs(z))

Associated confidence interval

\[ \text{C% confidence interval} = \bar{x} \pm z^\star \frac{\sigma}{\sqrt{n}}\\ \text{choose } z^\star \text{ s.t. area on standard normal distribution from }(-z^\star,z^\star)\text{ = C} \]

Minimum sample size to achieve margin of error:

\[ m^\star = z^\star \frac{\sigma}{\sqrt{n}} \iff n = (\frac{z^\star \sigma}{m^\star})^2 \]

If you lack access to the true value of the population standard deviation, a one-sample t-test for means is more appropriate.

If you want to compare means across two diferent independent populations, a two-sample z-test for means is more appropriate.

Considerations

A one-sample z-test conducted at confidence level \(\alpha\) will reject the null hypothesis if and only if the value corresponding to the null hypothesis, \(\mu_0\), is completely outside of the \(C = 1-\alpha\) confidence interval.