Thanks Paramita, this is a great and helpful tutorial!!!...

you are best

What to do when our data is yearly based...is it seasonal or nonseasonal

do you have git repo?

Hi @paramita, can you upload sarima.csv?

Thanks for the well explained theory

Hi its great content and knowledge sharing ! thanks ! please keep sharing more content like this

very good understanding of your expiation

thanks dor notes and data, where si the code ?

Thank you so much mam

thank you so much Paramita. Very well-explained.

One Que, My data is not stationary but as you mentioned i went with custom for loop to identify the p,d,q values and there d was 0 with lowest RMSE, but still data is not stationary so d should be one if i take diff by 1 , am i right? why that for loop suggests 0 value for d?

seasonal_decompose(df,model='additive',freq=4).plot(); this code may not be executes you can use period keyword instead of freq in that statement ()line 46 in original github code !

Hai how to use data in multiple sku along with sales date with two years

Dam Arima you look good 😍

Nicely and Perfectly Explained. Kudos to Paramita

can u give a full summary of machine learning explaing each M.L algorithm so that we can understand everything what involves in M.L

It seems that the data set that you provide has been corrupted. It contains information of just a month.

Clearly explained mam... Thanks alot

very helpful thanku

Finally, I have found a great teacher who can explain time series concepts with ease. It would be helpful if you could create a video on deploying machine learning models.

Won't we use SARIMA ? Given we are working on sales forecasting? This type of data has seasonality

Please Ma'am start to teach . Your content is very great.

Thank you :) It helps so much

Excellent video. Well explained & detailed.

Hello, Good day. If you can be of an assistance please. I working on a project work that has to do with forecasting using ARIMA. Can you please help me?

Thank you very much paramita..this video really helped me alot . practical implementation is what i was looking for. You deserve more ..thank you once again

please give solution ---- Input In [14] model=ARIMA(df.train,order=(pdq).fit() ^ IndentationError: expected an indented block

where is the video on acf and apcf plots

Your explanations are among the best. BTW... what about the SARIMA video? :)

mam why stopped posting videos. Its good

Many thanks to you. Great videos, very helpful!

Many thanks to you. Great videos, very helpful!

Amazing Lecture Mam

Hi Paramita, Very nicely explained tutorial. The csv that is provided has data only for January of the year 2014. Where can we see the rest of the data? Regards, KM

Sheer Brilliance. Won't ever forget how your channel helped me. May Lord Ram bless you 🙏 💜

Hello Paramita, thank you for explaining it so well. Just one note - in Seasonal decompose, frequency parameter has been deprecated and should be replaced with period parameter

Nice video, well explained, congrats and keep posting!

Thank you Ma'am great tutorial

Very useful tutorial. The best I've found on the net, thank you very much!

Thanks

The data is suitable for SARIMA/Holts Winter Method but you explained with ARIMA........!

You didn’t explain why to make a series stationary… 😠

you are brilliant ,please continue the rest of the parts out of 5

The iteratools method is outstanding. Thank you for sharing and congratulations for your talent.

Hii I am doing my data scientist course If you could provide more videos It will be a great help Or you can provide your notes plz

Thanks for video. I have some error : model=ARIMA(train,order=(5,0,4)).fit() ------ValueError: The computed initial AR coefficients are not stationary You should induce stationarity, choose a different model order, or you can pass your own start_params.

Hello Paramita, thanks a lot for your video. I wanted to ask you if you've read how to apply forecasting models to time series with multiple SKU (like 500 - 2000) considering the efficiency while running it, thinking of using the forecast once every week. I would really appreciate if you can indicate me a study case or real case in which I can take a look at the approach within the code. Thanks in advance!!

More than Great😍

Thank you so much Ma'am but can you also explain how to do the hourly prediction (24 hrs). I would be helpful if you explain it.