CFA Level I Study Guide: Hypothesis Testing
Hypothesis testing is one of the highest-weighted Quantitative Methods topics in CFA Level I.
It is used to determine whether sample evidence is strong enough to support or reject a claim
about a population.
1. Core Definition
Hypothesis testing is a statistical procedure used to determine whether there is enough evidence
from a sample to reject a claim (hypothesis) about a population parameter.
Instead of proving something is true, hypothesis testing evaluates whether the sample evidence is
strong enough to reject the null hypothesis.
2. Simple Explanation
Imagine a company claims:
"Our batteries last 10 hours."
You buy 50 batteries and find the average life is 9.3 hours.
Now you ask:
"Is this difference simply due to random chance, or is the company's claim likely false?"
Hypothesis testing answers this question.
Steps in Hypothesis Testing
The CFA Institute expects you to know these six steps.
Step Action
1 State the null hypothesis (H₀)
2 State the alternative hypothesis (Hₐ)
3 Choose the significance level (α)
4 Calculate the test statistic
5 Compare with the critical value (or use the p-value)
6 Reject or fail to reject H₀
Important Symbols
Symbol Meaning
μ Population mean
x̄ Sample mean
σ Population standard deviation
s Sample standard deviation
n Sample size
α Significance level
β Probability of Type II error
df Degrees of freedom
Significance Level (α)
The significance level is the probability of rejecting a true null hypothesis.
Common values:
Confidence Level α
90% 0.10
95% 0.05
99% 0.01
Confidence Interval vs. Hypothesis Test
These are closely related.
If the hypothesized value lies inside the confidence interval, fail to reject H₀.
If it lies outside the confidence interval, reject H₀.
Test Statistic
A test statistic measures how far the sample statistic is from the hypothesized population value,
in units of standard error.
The larger its absolute value, the stronger the evidence against H₀.
t-Distribution
The t-distribution is similar to the normal distribution but has fatter tails.
Use it when:
Population standard deviation is unknown.
Sample size is small (especially (n<30)).
As sample size increases, the t-distribution approaches the standard normal distribution.
Chi-Square Distribution
Characteristics:
Used for variance tests.
Never negative.
Right-skewed.
Shape depends on degrees of freedom.
Degrees of Freedom
F-Distribution
Characteristics:
Used to compare two variances.
Always positive.
Right-skewed.
Depends on two degrees of freedom.
PART 5: Test of Correlation
Purpose:
Determine whether the population correlation differs from zero.
Hypotheses
where:
(\rho) = population correlation coefficient.
Test Statistic
Components
(r) = sample correlation coefficient.
(n) = sample size.
Use the t-distribution.
Degrees of freedom:
df= n-2
Degrees of Freedom (df)
Degrees of freedom represent the number of independent pieces of information available for
estimating a parameter.
4. Step-by-Step Solving Guide
A. Single Mean Test
Step 1
State:
H₀
Hₐ
Step 2
Choose the significance level ((\alpha)).
Step 3
Determine whether to use:
Z-test (σ known).
t-test (σ unknown).
Step 4
Calculate the test statistic.
Step 5
Find the critical value (or compare the p-value with (\alpha)).
Step 6
Decision:
Test statistic in rejection region (or p-value < α) → Reject H₀.
Otherwise → Fail to reject H₀.
B. Variance Tests
1. Identify whether one or two variances are being tested.
2. One variance → Chi-square test.
3. Two variances → F-test.
4. Compute the test statistic.
5. Use the appropriate distribution and degrees of freedom.
C. Correlation Test
1. Compute the sample correlation (r).
2. Calculate the t-statistic.
3. Use (df=n-2).
4. Compare with the critical t-value (or p-value).
5. Exam Tips for Difficult Questions
1. Always Start with the Hypotheses
Read the wording carefully:
"Equal" → H₀.
"Different" → Two-tailed Hₐ.
"Greater than" → Right-tailed Hₐ.
"Less than" → Left-tailed Hₐ.
4. Don't Confuse Type I and Type II Errors
Error Meaning Probability
Type I Reject a true H₀ (\alpha)
Type II Fail to reject a false H₀ (\beta)
5. Remember Statistical Power
Higher power means a greater chance of detecting a false null hypothesis.
6. Confidence Intervals and Hypothesis Tests Are Equivalent
At the same confidence level:
Hypothesized value inside the confidence interval → Fail to reject H₀.
Hypothesized value outside the confidence interval → Reject H₀.
Final CFA Level I Takeaway
Hypothesis testing follows a consistent framework: state H₀ and Hₐ, choose a significance level,
calculate the appropriate test statistic, compare it with the critical value (or use the p-value), and
make a decision. The key to success on the CFA exam is recognizing which parameter is being
tested, because that determines the correct distribution:
Z → Single mean (known population standard deviation).
t → Means and correlation.
Chi-square → One variance.
F → Two variances.