CODEWITHSUNDEEP
time-seriesarimasarimaxpmdarima
Divyansh Sharma
Divyansh Sharma

Reputation: 31

Issue with SARIMA model for PM10 concentration forecasting with m=365

I'm trying to build a SARIMA (Seasonal Autoregressive Integrated Moving Average) model for forecasting PM10 concentrations based on five years of data. However, when I set the seasonal parameter m to 365, my code doesn't seem to run.

Could someone please explain why my code is not running with m=365 and suggest a potential solution?

Thanks in advance!

Here's a snippet of my code:

Split the dataset into train and test sets

`train_size = int(len(Alipur_df) * 0.8)  # 80% train, 20% test`
`train, test = Alipur_df[:train_size], Alipur_df[train_size:]`

Convert train DataFrame to a numpy array

`train_values = train['Alipur'].values`
`test_values = test['Alipur'].values`

Use auto_arima to find the best parameters for SARIMA

`auto_model = auto_arima(train['Alipur'], seasonal=True, stationary=True, m= 365, trace= True) `

Upvotes: 0

Views: 22

Answers (0)

Related Questions