ARIMA in python. Best way to Identify p d q. Time Serie Forecasting. With Example. Free Notes.

[@ratheeshmsuresh7368]

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.

[@AiykRichie]

I agree with teaching how to get this deployed.

[@cogcog312]

Simply excellent. Straight-forward, concise, well-explained and detailed. Thank you! You need to do more videos as you seem to have a natural talent to teach. Not everybody has it.

[@yogeetabhoyroo5934]

Please advise on the title of ACF pacf video

[@namrathakaranth]

This the best video I have come across. Bless you maam

[@haridaasan]

Madam this is the best!!! Quite underrated i would say! A great video and thanks a million for clarifying the pdq selection. Almost everyone talked about pacf and acf and everyone seemed to have their own way of telling how to do it - which was confusing. The custom for loop is the best i have seen.

[@dganesh3641]

First class teaching, very nice, clear and attention grabbing

[@MUNIKUMARNM]

One of the best vedio availablel in youtube for ARIMA

[@bunkoti]

Ma'am, this is the best ARIMA explanation I have come across on KZbin. Can you please make videos on SARIMA and SARIMAX as well, along with other ML algorithms? You truly deserve to have a lot more subscribers. Thanks.

[@CoachTobby]

I really need her videos on this

[@jesusaanaya5625]

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

[@kvafsu225]

Brilliant Madam. So clear, even a novice can understand.

[@bumpagab]

I really appreciate your teaching style. ! Thank you so much for great content.

[@sohamjondhale7183]

Tysm ma'am recently ARIMA model was updated because of which I was having more problem in forecasting. I already spent 2 days forecasting my model but it always gave me some or the other error. When I saw your video in couple of hours i forecasted my dataset. Tysm once again ma'am for ur methodology 🙏

[@ngoclinhnguyen5439]

thank you so much Paramita. Very well-explained.

[@fernandoyanez9891]

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

[@jeffwong1310]

Thank you so much for this video, it helps me to build my ARIMA model. I like your alias: Paramita. You definitely have the "Prajna"!

[@davidajekigbe9828]

Many thanks for this your video on ARIMA. It is a great one.

[@sumertheory]

Best Arima video on youtube! 😀

[@faroozrimaaz7092]

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

[@nujanai]

Excellent video. Well explained & detailed.

[@Nikhil-hi1qs]

Best explanation on ARIMA

[@amitjain000]

very good description, appreceate your teaching skill

[@bonganindlovundlonu3239]

Thank you for super explanation. This is the best.

[@Adinasa2]

very good presentation , very useful and helpful

[@HiltonFernandes]

Thanks a lot for a great video, and for sharing data and presentation.

[@ct3326571886]

Madam it will be immensely beneficial if you kindly explain that since the data used here was non-stationary, was it not necessary to convert the data into a stationary one before feeding it to a machine learning model? if so, if you kindly care to explain. Excellent Video by the way. Really thank you so much for the beautiful explanation. Sincere Regards

[@datascienceandaiconcepts5435]

U deserve more subscribers, Good Explanation

[@prakash.penterpreneur6166]

very good understanding of your expiation

[@ahmadqureshi1305]

Hi Paramita, this is extremely insightful, thank you! Would you be able to share the notebook too? Thanks again!

[@flashretry317]

Amazing Lecture Mam

[@Mukeshkumar-yl1qq]

Truly deserves lot more subscribers 👏 🙌 💖

[@deepansinha7687]

So good video. I think this video sums up all theories very well

[@PatricioStegmann]

Nice video, well explained, congrats and keep posting!

[@shubhammaurya492]

Thank you Ma'am great tutorial

[@Denis-fd5kr]

Many thanks to you. Great videos, very helpful!

[@parthamukherjee6944]

Thank you for creating this video! Super helpful!

[@rluijk]

Really good explanation and overview! Showing mastery and practical use.

[@paramita2674]

Thank you ..

[@amitmunu]

Excellent explanation. Kudos!!

[@milindwaghmare2780]

amazing very well explained

[@nathanthreeleaf4534]

I was hoping you'd go into more detail about the seasonality aspect of the data and dealing with the seasonal_order parameter of the ARIMA function. Would it work the same way to create product sets for the P, D, Q and S values and sending them into the model to test for the lowest MSE? Or do you have another video that touches on that further perhaps? All that aside, this was a great video and really helped me work through this process step-by-step.

[@jongcheulkim7284]

Thank you. This is very helpful.

[@inesdhaker2476]

Excellent explanation !!!!!

[@paramita2674]

Thank you...

[@analyticsWithJay]

Thank you, this tutorial is really good, would like to see many videos, Cheers

[@nimishghule]

Great video maam.

[@pranavbapat1909]

Very nice explanation... superb.

[@paramita2674]

Thank you..

[@sharmilasenguptachowdhry509]

Thank you, waiting for your SARIMA lecture

[@paramita2674]

Thank you..

[@legendtrevor7284]

i guess I am kinda randomly asking but does anybody know of a good place to stream new movies online?

[@lanehassan6017]

@Legend Trevor i watch on Flixzone. Just search on google for it :)

[@tristanpaul2116]

@Lane Hassan yup, I've been watching on Flixzone for months myself :D

[@legendtrevor7284]

@Lane Hassan thank you, I signed up and it seems like a nice service :) I really appreciate it!

[@enomis9478]

Thank you for this great tutorial. However, I did not understand a point. Why did you choose d = 0? In your initial analysis you showed that the series was non-stationary. Therefore, to build the correct model it would be necessary to differentiate at least 1 time, i.e. d = 1.

[@rafaelfraga7976]

I am thinking the same, this choose of p, d and q is a little bit strange because after setting as stationary we should use d = 1

[@shreyanshgaurkar9107]

actually if we see that if the time series is already stationarity then we dont want to differencing we directly get the value d= 0 but if the time series is not stationarity then we can differenciate these by 1st order differenciation to make the time series stationarity so due to first order differencing we get the value d = 1

[@fuzailkhan9027]

Great Explanation, really helpful. can you please share the link for the video for PACF and ACF plot and how to determine the p d q values from those charts

[@adityachopra5688]

the dustbin animation was spot on

[@davidfullstone]

Great tutorial, thanks.

[@_AbUser]

""We will not talk about bookish theory coz it has no any practical implementation" - One the the useful things that should to say at the start! That 100-true but nobody talking it.. )))

[@karthebans2420]

It's very good explanation. Can you please make the video on SARIMA and other time series algorithms like Prophet, ThymeBoost, LSTM etc.,

[@udayshuklabcp2782]

very helpful thanku

[@chautrannguyentran9408]

Hi, I have one question How to used ARIMA if we have multi variables? For example, Y= sales X1=laptop , X2=TV, X3= newspaper, X4=radio, and X5=cellphone

[@rdeepti5583]

SIMPLY SUPERB!!!

[@AISteveMusic]

Thanks a lot for your videos, they are to the point and easy to follow. I hope you continue to develop this youtube channel! Only thing that could be better is the audio quality :)

[@pranavbapat1909]

Although it would be really nice if you make some more videos on time series analysis on univariate and multivariate data, and also using XGBoost, Linear Regression, Random Forest, Simple Exponential Smoothing, and so on... And a video explaining which method to use when for what type of data. :)

[@hectorg.m.3350]

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

[@khairulfahim]

Your video is quite good. Please make a full playlist on Time Series Analysis.

[@arpitsrivastava7559]

Great video. Can you provide a bigger dataset? This one has only 32 rows.

[@vladk9152]

Looking forward for a SARIMA video

[@pratyushkumar8174]

Wooww yr...too gud

[@anfalbilal1985]

The best video about arima model. Thank you very much. Can you send the Link of the video about acf and pacf that you mentioned at the end. I searched on your channel and I didn't find it. I am waiting your replay..

[@arungireesh686]

you are best

[@adeolasolomon7514]

SIMPLY THE BEST

[@siddheshmhatre2811]

Thank you so much mam

[@jeffwong1310]

I wonder whether you will have the sharing of running a SARIMA model instead

[@poojabairagi7018]

thanku😇😇

[@priyaarora4436]

Hey you have not uploaded videos on PACF, and ACF. Also, why have you stopped creating videos. You genuinely explain very conceptually unlike the famous ones who themselves are confused, but still have 541k subscribers!

[@davidgyang1575]

True. She explains better than most of videos I have watched

[@HealthyFoodBae_]

So good 🙏🏽

[@paramita2674]

Thank you..

[@krishcp7718]

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

[@howcouldi4974]

Brilliant

[@anjujagadish2739]

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.

[@goncaloalmeida1451]

Before doing the arima model, shouldn't you make the time serie stationary??

[@isaacdare4195]

Was just wondering

[@vikaskatoch2454]

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

[@luckyytb]

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.

[@adilmajeed8439]

where i can find the data that you have used in the video? The github reference doesnt contain the reference file while loading into the dataframe

[@kunalpandya8468]

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?

[@instasol2453]

Could you pls let me know where is the video for judging p and q values from ACF and PACF plots?

[@amitsuryawanshi8632]

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

[@MusicMonster26000]

This was great, can you do a SARIMA walkthrough?

[@mingkwan5280]

I dont have any background in a programming language.. I have a problem.. Do you just insert all these code in one file or it has to be separated? and if it has to be separated then which code should be on the same file?

[@tac3523]

Dam Arima you look good 😍

[@soniayadav9804]

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

[@stonesupermaster]

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!!

[@arunthandra5065]

Great explanation. Can you please provide the code...

[@anantkumarnagar7397]

did you uploaded the video of seasonal arima

[@ashwin_.0710]

Shouldn't the dataset be made stationary before proceeding with the modeling? If not what was the point of checking stationarity? Or does the d parameter automatically do the job? PLEASE EXPLAIN DIFFERENCING/BOX COX TRANSFORMATION TO MAKE DATA STATIONARY !!!

[@Stevejobs6343]

mam can you please share that Jupyter notebook, it is really needed for my project

[@2dapoint424]

Can you share/ upload the python notebook to your github link?

[@Gear2Gaming]

Have you deleted the acf pacf plot video? and Sarima as well

[@dzandulawrence4018]

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?

[@joeljose41]

Mam how to know when to use multiplicative or additive in decomposition

[@ravindarmadishetty736]

Hi @paramita, can you upload sarima.csv?

[@oghbazghikafel708]

thanks

[@geetikapanda8835]

Can you please share the link for SARIMA?

[@_IT_Jason]

Can you explain how you got pdq values

[@sriramram2166]

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

[@pouriaforouzesh5349]

👍

[@kartiksharma-yw7qf]

Ma'am am working with amazon stock price,so m I suppose to resample it to MS or I should something like 'B'?

[@paramita2674]

Stock price is given and generally analysed on a daily basis so use ‘B’

[@kartiksharma-yw7qf]

@@paramita2674 but while using B there are some 0 on some dates so is it fine to ffill on those dates as it will decrease the pace of model .

[@ejelonubenedict8270]

Please where's the link to the Seasonal Arima lecture?

[@AI_Aficionado]

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