Null Hypothesis

What Is a Null Hypothesis?

A null hypothesis is the starting assumption in a hypothesis test, a statistical procedure used to judge whether observed data reflect a real pattern or simply chance. It usually states that there is no effect, no difference, or no relationship. For example, it might say that a trading strategy earns the same average return as the market, or that two groups of stocks perform equally well.

The null hypothesis is often written as H0. It is paired with an alternative hypothesis, which states what the researcher suspects might be true, such as that the strategy does earn a different return. The test does not try to prove the alternative directly. Instead, it asks how surprising the data would be if the null hypothesis were true.

How It Is Used

Analysts collect data, then calculate a test statistic and a p-value. The p-value estimates the probability of seeing results at least as extreme as the observed ones, assuming the null hypothesis holds. A small p-value suggests the data are hard to explain under the null, which is treated as evidence against it.

Before testing, analysts choose a significance level, a threshold for how small the p-value must be before the null hypothesis is rejected. A commonly used threshold is 5%. If the p-value falls below it, the result is described as statistical significance. If not, the analyst fails to reject the null hypothesis, which is not the same as proving it true.

Why It Matters to Investors

Investors often encounter null hypotheses when reading research claims, such as a backtest showing that a strategy beat the market or a study linking a signal to future returns. Asking what the null hypothesis was, and whether the test controlled for type I error (a false positive) and type II error (a false negative), helps judge how much weight a result deserves.

Rejecting a null hypothesis does not guarantee a result will persist. Markets change, many tests run on the same data can produce false positives by chance, and statistical significance does not measure economic importance. Careful investors treat statistical findings as one input among many rather than as proof.