Study guides / CFA L1: CFA Program Level I / Sampling & hypothesis testing
Sampling & hypothesis testing
CFA L1: CFA Program Level I · quick review
Drawing inferences from samples: sampling methods, the central limit theorem, confidence intervals and hypothesis tests, including errors and p-values.
The rules
- The central limit theorem: for large samples (n ≥ 30), the sample mean is approximately normal with standard error σ ÷ √n, whatever the population's shape.
- Sampling biases include survivorship, look-ahead, time-period, data-snooping and sample selection bias.
- A confidence interval = point estimate ± reliability factor × standard error.
- Hypothesis test steps: state H₀ and Hₐ, choose the test statistic and significance level, set the decision rule, compute, decide, and interpret.
- Type I error: rejecting a true null (probability = significance level α). Type II error: failing to reject a false null. Power = 1 − P(Type II).
- The p-value is the smallest significance level at which the null can be rejected; reject H₀ if p < α.
- Use a t-test for means with unknown population variance, chi-square for a single variance, F-test for equality of two variances.
Formulas
- Standard error of the mean: σ ÷ √n (or s ÷ √n)
- Confidence interval: x̄ ± z(or t) × SE
- Test statistic (mean): (x̄ − μ₀) ÷ (s ÷ √n)
Common traps
- Failing to reject H₀ doesn't prove it true.
- Lowering α reduces Type I errors but raises the chance of Type II errors, for a given sample size.
- A larger sample shrinks the standard error and narrows confidence intervals.
How it's tested
Choosing the right test, interpreting p-values and errors, and computing standard errors or intervals.
Practise it
The full lesson, worked examples and practice questions are in Tickmark, where every answer choice is explained.
Open the full lesson