Why augmented dickey fuller test gives me less p-value i.e. less than 0.05 despite of having seasonality my data.

My data's plot:

enter image description here

Code for adf test :

from pandas import Series from statsmodels.tsa.stattools import adfuller

X = nottemts
result = adfuller(X)
print('ADF Statistic: %f' % result[0])
print('p-value: %f' % result[1])
print('Critical Values:')
for key, value in result[4].items():
    print('\t%s: %.3f' % (key, value))

The output I get is,

ADF Statistic: -3.240917 p-value: 0.017735 Critical Values: 1%: -3.460 5%: -2.874 10%: -2.574

This test is used to check the stationarity i.e. Mean, Variance, autocorrelation structure(Seasonality) doesnot vary with time.

Thanks in advance.

  • $\begingroup$ The dicky-fuller test is used to test if time series is stationary. Which based on your graph, just might be. $\endgroup$ – yoav_aaa Jul 24 at 14:35
  • $\begingroup$ But it has seasonality present in it i.e. the autocorrelation changes over time. Right ? Hence I guess it is Non-Stationary Series. $\endgroup$ – Raja Sahe S Jul 24 at 14:41
  • $\begingroup$ Yes, you are right by definition, series should not be considered stationary. There is a concept of Cyclostationary which seems to fit your series. See this related post. $\endgroup$ – yoav_aaa Jul 24 at 15:00
  • $\begingroup$ Can you please provide the related post. $\endgroup$ – Raja Sahe S Jul 24 at 15:22
  • $\begingroup$ Yes - - stats.stackexchange.com/questions/131092/… $\endgroup$ – yoav_aaa Jul 24 at 15:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.