Mann-Kendall trend test in STATA (Tutorials)

[Ramon Rodriguez-Santana]

Hello Everyone,

I am trying to find out how to do a non-parametric Mann-Kendall trend test to detect monotonic trends.

Here is some example data:

Year variable values = 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018
Events variable values = 420, 350, 305, 288, 250, 209, 175, 89

Thanks!

-RRS

[German Rodriguez]

Hi Ramon Rodriguez-Santana. The Mann-Kendall trend test is based on the Kendall tau correlation between the outcome and the time variable, so you can use Stata's -ktau- command. Here are your data using -dataex- as recommended:

* Example generated by -dataex-. To install: ssc install dataex
clear
input int(year events)
2011 420
2012 350
2013 305
2014 288
2015 250
2016 209
2017 175
2018  89
end

And here is the result of the test:

. ktau year events

  Number of obs =       8
Kendall's tau-a =      -1.0000
Kendall's tau-b =      -1.0000
Kendall's score =     -28
    SE of score =       8.083

Test of Ho: year and events are independent
     Prob > |z| =       0.0008  (continuity corrected)

You have clear evidence of a negative trend over time. Results agree exactly with the MannKendall function in the R package Kendall.

[Ramon Rodriguez-Santana]

Hi German Rodriguez,

Thanks for all the information... especially for the STATA Kendall tau correlation example.

Sincerely,

-RRS