Learning notebook Part 1 - Sarimax parameters

LDSSA / batch2-BLU03

Predict timeseries in the real world

MIT License

0 stars 4 forks source link

Learning notebook Part 1 - Sarimax parameters #3

Closed antoniopleite closed 4 years ago

antoniopleite commented 6 years ago

Although my best_model_params are (1,1,1,0,1,1,14) the example below that optimization fits a model, claiming that it's about to use the best values, with (0,1,1,1,1,1,12) parameters. What am i missing? Some kind of randomness?

pma2 commented 6 years ago

Any comments about this?

deeproute commented 6 years ago

Yea wondering the same thing.