This presentation discusses and illustrates the basic principles of ARIMA modelling for forecasting a non-seasonal (or seasonally adjusted), time series.
Very insightful and well structured video but ones you applied the differentials and the afc showed some of the data being unstationary ...you suggested an ARIMA mode of (3,1,0)...whats the bases of that suggestion
where does splitting data into train and test sets fit into this ? I thought we only select the model with lowest Aicc or BIC based on how the training set performs on the test set?
I don't think that approach is appropriate with time series data because the observations are not independent and you can't simply extract some of the data into different sets or you would destroy the integrity of the data.
It doesn't really matter which signs are used in this general specification of the ARIMA model. Some authors/texts use minus signs as I have done here, others use plus signs. When a model is estimated by the software package the appropriate signs on the coefficient estimates will be determined, and those will be used in the equation that will be used for forecasting.
Nice lecture sir. Thanks but i have a doubt and wishes to share with you. Can you please share your email id to share the problem in ARIMA modelling, I face.
Bravo 👏, great video . Thanks a lot sir for sharing your knowledge 😊
Such a well articulated and to the point video . Really deserves more views . Thanks
very useful!!!!!!!!!!!!!!!!!!!! Love you
very clear and logical explanation! Thank you
such a great guider about the data analysis , like it 😍
Crackin' presentation. I got a lot out of that, thank you!
Your voice is so wonderful, thank you for the explanation.
Great video, thank you.
very useful video, thank you!
Simple explanation.thanks
Nice presentation.
Very insightful and well structured video but ones you applied the differentials and the afc showed some of the data being unstationary ...you suggested an ARIMA mode of (3,1,0)...whats the bases of that suggestion
thank you sir it was very useful 👍
What is name of the book from where example is taken?
a very nice and warming voice
where does splitting data into train and test sets fit into this ?
I thought we only select the model with lowest Aicc or BIC based on how the training set performs on the test set?
I don't think that approach is appropriate with time series data because the observations are not independent and you can't simply extract some of the data into different sets or you would destroy the integrity of the data.
what happens if at 1st difference all the data are not significant
many resource i read show that the sign (-) in the MA equation actually sign(+) . now i am confused can you explain it sir :)
It doesn't really matter which signs are used in this general specification of the ARIMA model. Some authors/texts use minus signs as I have done here, others use plus signs. When a model is estimated by the software package the appropriate signs on the coefficient estimates will be determined, and those will be used in the equation that will be used for forecasting.
Nice lecture sir. Thanks but i have a doubt and wishes to share with you. Can you please share your email id to share the problem in ARIMA modelling, I face.