Most of the significance tests based on Z, t and F distributions make use of the assumption that the sample drawn from the population is random in nature. Run test is used to test the randomness of the sample selected from the population. A ‘run’ is a sequence of repeated occurrence of a particular symbol.
For instance, there are seven runs in the following sequence of M and F.
MMMM FFF M FFFF M FFFF MMM
Run test of randomness is particularly applied in the stock market to know if the stock price of a particular company behaves randomly or follow any pattern. The hypotheses for this test are:
Null hypothesis-H0: The sample is random in nature.
Alternative Hypothesis- H1: The sample is not random.
TYPES OF RUN TEST
- Run Test with Interval or Ratio Scale Measurement.
2. Run Test with Nominal Scale Measurement.