Time Series Exponential smoothing | Exponential smoothing in time series-must know concept

Time Series Exponential smoothing | Exponential smoothing in time series-must know concept
Hello,
My name is Aman and I am a data scientist.
About this video:
In this video I explain about exponential smoothing of time series. I explain how exponential formula works and what are the different formula for time series exponential smoothing. I also explain the concept of additive and multiplicative time series in this video. Below topics are explained in this video:
1. Exponential smoothing in time series
2. Types of time series exponential smoothing
3. Additive and multiplicative time series
4. Use of time series smoothing
5. Exponential smoothing in python
About Unfold Data science: This channel is to help people understand basics of data science through simple examples in easy way. Anybody without having prior knowledge of computer programming or statistics or machine learning and artificial intelligence can get an understanding of data science at high level through this channel. The videos uploaded will not be very technical in nature and hence it can be easily grasped by viewers from different background as well.
Join Facebook group :
groups/41022...
Follow on medium : / amanrai77
Follow on quora: www.quora.com/profile/Aman-Ku...
Follow on twitter : @unfoldds
Get connected on LinkedIn : / aman-kumar-b4881440
Follow on Instagram : unfolddatascience
Watch Introduction to Data Science full playlist here : • Data Science In 15 Min...
Watch python for data science playlist here:
• Python Basics For Data...
Watch statistics and mathematics playlist here :
• Measures of Central Te...
Watch End to End Implementation of a simple machine learning model in Python here:
• How Does Machine Learn...
Learn Ensemble Model, Bagging and Boosting here:
• Introduction to Ensemb...
Build Career in Data Science Playlist:
• Channel updates - Unfo...
Artificial Neural Network and Deep Learning Playlist:
• Intuition behind neura...
Natural langugae Processing playlist:
• Natural Language Proce...
Understanding and building recommendation system:
• Recommendation System ...
Access all my codes here:
drive.google.com/drive/folder...
Have question for me? Ask me here : docs.google.com/forms/d/1ccgl...
My Music: www.bensound.com/royalty-free...

[@AnujKinge]

One of the most underrated channels for Data Science. The channel truly deserves a million subscribers

[@UnfoldDataScience]

Thanks Anuj. Your comments are my motivation

[@rameswarmukherjee7404]

Please make more video in time series course, so people like us can understand simple way to make prediction in our workplace

[@WenPan]

Thank you very much for this clear and concise explanation of exponential smoothing.
Wish you all the best :)

[@jijopvarkey]

Such a concept was explained so well..
Thank you..

[@alphaq721]

I have a doubt sir. Do we choose Alpha value based on our wish or is it determined out of something?

[@AacharyaVikash]

superb explanation.. thanks a lot

[@moodysaleh1721]

I used Neural Prophet to forecast a time series for electric load usage. One of the options, asks whether I want to use holidays as a feature and whether they should be additive or multiplicative. Knowing that during holidays and weekends, the electric usage drop to the minimum level, which one do you think makes more sense, additive or multiplicative for holidays/weekends? I am assuming additive, especially since the whole average of the time series is not increasing

[@jambukeshwara1844]

Really thanks bro ❤❤

[@akshayingole6442]

Thanks a lot sir you made this tough concept so easy to understand that i will remember it very effectively without any confussion

[@UnfoldDataScience]

Welcome Akshay.

[@aliabbasrizvi5431]

Thanks Sir very well explained

[@vanathiuthaman4861]

Nice ecplantion sir

[@mohdravi8541]

Excellent video

[@Motivational_Greek]

Great,

[@murphyslaw2366]

Dear Amant,
Very nice explanation.
Do you have any video that explain the below topics?
1. Cover windowed/convolution smoothing
2. Lowess
As i am a beginner, it will be helpful if you have any sample python code for Convolution, Lowess and exponential smoothing for practical understanding.

[@MrRynRules]

Thank you sir!

[@meysamjavadzadeh]

tnx man really helpful and joyful explaining keep making videos 👌💣❤

[@TheOraware]

Very nice explanation , does exponential smoothing work on few number of records? Suppose I have 5 years student enrollement data from 2018 to 2022 and want to forecast number of enrollment for 2023, can it be achievable by exponential smoothing? There is only yearly data available not monthly , not weekly/daily. After every end of year we have how many students enrolled this year and based on these previous and current years we want to forecast number of students before starting next year

[@farogatolimjonova9566]

Thank you a lot! It is much more understandable now

[@UnfoldDataScience]

Glad you liked it.

[@rajeshmane2486]

It is great to watch this video, thy way you explain is amazing.

[@UnfoldDataScience]

Thanks Rajesh 🙂

[@user-nj2ze7jw6v]

Dear Aman, How about the 'Cyclical' component. What are your thoughts on that.

[@NitishKumar-zd9ej]

Thank you Aman for the explanation.
One doubt with respect to Triple Exponential Smoothing: You have used gamma to compute C(t). I believe the value of C(t) = gamma times (x[t]/y[t]) + (1-gamma) times C(t-l) . However, you have written in the video (1-alpha) in stead of (1-gamma) while multiplying with C(t-l). Please clarify whether the 2nd multiplication should have (1-alpha) or (1-gamma). Thanks

[@yoyomovieclips8813]

yes nitish,it should be 1-gamma in the third equation which captures seasonality.

[@UnfoldDataScience]

Thanks Both of you Neeraj and Nitish. should be 1-gamma. My mistake it seems.

[@ajaybhatt6820]

hi aman , what is xt in third exponential??

[@lovleenkaur1042]

Great work... Can u recomend a book for it?

[@mattym4105]

Thank you very much! Cheers from Toronto 😎

[@UnfoldDataScience]

Cheers Matty.

[@Theviswanath57]

clear explanation thanks man

[@UnfoldDataScience]

Thanks Viswa

[@Ramen8777]

Wow

[@krishnabarfiwala5766]

Very nicely explained.

[@UnfoldDataScience]

Thanks for liking Krishna.

[@padmaprasad1250]

Thank you for the explanation. Quick question, Does exponential smoothing apply to compositional data?

[@UnfoldDataScience]

Yes it does!

[@sandipansarkar9211]

finished watching

[@mixysujith8309]

Sir
Very much useful.
Excellent presentation.
Can i get explanation for time series anomaly detection with examples and some models

[@UnfoldDataScience]

Will upload soon

[@aruncn123456789]

Very nice!!

[@UnfoldDataScience]

Thank you Arun!

[@hiteshyerekar9810]

Good Video Aman.

[@UnfoldDataScience]

Thanks Hitesh.

[@subhasisdutta3711]

when do we get negative predictions and what's the reason for negative forecast basically in time series ?

[@UnfoldDataScience]

It depends on what training and test data we are giving.
For example 5,4,3,2,1,0,-1,-2,-3 then next prediction may be -4

[@poojapatil3767]

Well explained

[@UnfoldDataScience]

Thanks pooja

[@sudheeshe1384]

Good explanation

[@UnfoldDataScience]

Thanks for liking

[@venkateshsadagopan2505]

Nice video. What will be the initial values of y_0 , b_0. Is it 0 ?

[@UnfoldDataScience]

small positive number.

[@venkateshsadagopan2505]

@@UnfoldDataScience ok thanks

[@likithaguntha8105]

Is it possible to build a model when there is no trend noe seasonality

[@UnfoldDataScience]

YEs

[@saneilshah9855]

Amant the goat

[@abhishekgaurav7786]

sir what is residuals
please explain about it

[@UnfoldDataScience]

Hi Abhishek, residual means error.

[@abhishekgaurav7786]

@@UnfoldDataScience sir are you on linkdin
can you please explain about all the test
like t test z test ,p value adf,anova

[@TheNishi42]

Not sure if this playlist is in sequence ??

[@UnfoldDataScience]

I will Check Romita.

[@aashitaburman9637]

Sir, I have a doubt. How do we calculate beta and gamma?

[@UnfoldDataScience]

These are parameter we need to find for specific data

[@yoyomovieclips8813]

Sir, i read in a book where moving average,weighted moving average,single/double/triple exponential smoothing techniques have been used as a model to do the forecasting itself.So does it mean that all these techniques can be used to do the forecasting directly by applying these oris it that these techniques can only be used to do smoothing of the time-series data?

[@UnfoldDataScience]

both ways we can use these techniques.

[@muhammadmasudtarek7397]

last equation for Ct; is it (1-alpha) or (1-gamma)? just want to be sure.

[@muhammadmasudtarek7397]

ok I got it. It should be (1-gamma)

[@smaranrai4898]

If a company only has past data for 1 year, how should it forecast for future months that will mitigate the seasonal effects?

[@UnfoldDataScience]

Not possible to forecast if there is seasonality in business. Some simpler techniques can be used to get a rough estimate.

[@smaranrai4898]

@@UnfoldDataScience thankyou for responding. can u please name some of those simpler techniques.

[@numnum6708]

nice lecture but pls writing on board shud be dark

[@UnfoldDataScience]

Thanks, feedback taken.

[@asareosbornpeprah7201]

Sir please what about the cyclic component of time series?

[@HARSHRAJ-2023]

Nice video. If you correct your pronunciations then you will go a long way( e.g. Its Zero not Jiro and many more).

[@UnfoldDataScience]

Thanks again Harsh. I ll work on it.

[@sandipansarkar9211]

finished coding

[@RaghavStar6]

Dear, Sir.
In English we can learn by ourselves from book. So if you are making the video so make in Hindi language also because we are INDIAN ❤. FROM next time make the videos in Hindi language and try you best to explain in hind.