About training ARIMA

AI & Machine learning
1

Course Name: Financial Time Series Analysis for Trading, Section No: 18, Unit No: 11, Unit type: Notebook

Hello, why in this unit we're training the model indicating arguments to the model? In AR and MA we never indicated nothing in model.fit(), however in this unit we're indicating such parameter. What's the difference about including start_params in model.fit()?

https://i.imgur.com/ExqocKU.png

 

2

And see what we did for AR and MA models:

3

Hey Daniel,



We will check this and get back to you.

4

Okay, thanks.

5

Hi Daniel,



The start_params parameter in model.fit() provides a way to customize the initial parameter values for the optimization process, potentially improving the efficiency and effectiveness of training machine learning models. If start_params is not present then the initial parameters of the model are randomly initialized.



For the sake of simplicity and to enable better understanding in the initial notebooks, start_params was not added in AR, but you can choose to do this. Doing so will allow you to initialize the model with specific parameter values rather than using default initialization.



Hope this helps!



Thanks

Rushda

6

Very clear, thank you!