Mann-Kendall Test for Trend

See Also:

Description:

The Mann-Kendall trend analysis is provided by Gilbert (1987) as a non-parametric test for trends in data over time. The test is suitable for data that do not follow a normal distribution, and supports multiple observations per time period. There is no physical limit on the data set size. Only testing of individual monitoring locations for trends is supported. Testing groups of monitoring locations is not supported. The test is performed at a confidence level selected by the user.

Use:

To detect the presence of an upward, downward, or two-tailed trend in data over time.

Remarks:

ChemStat supports this test for tied values, and multiple observations per time period. Analyzing data from multiple wells simultaneously is not supported.