Time Series Forecasting Theory Part 1 - Datamites Data Science Projects

[@lucienkalmogo3448]

thank you so much !! concise, short and very clear

[@DataMites]

Glad you liked it!

[@shamilsons]

Clear and good explanation of Time Series Analysis concepts. Thank you for sharing information

[@DataMites]

Thank you please watch part 2 video here kzbin.info/www/bejne/emrcZ5eZrpWrgZo

[@murtalaibrahim8352]

am always following your nice and wonderful programmes, please keep on informing us anytime you have a training session many thanks

[@kc_good_luck]

It's the best time series tutorial I've watched so far!! Thank you!!

[@DataMites]

Most welcome!

[@TheMunishk]

You really explained well. With a clear voice and thoughts.

[@DataMites]

Thank you, Please watch Time Series Forecasting Theory Part 2 kzbin.info/www/bejne/emrcZ5eZrpWrgZo

[@k_pro4110]

This is amazing. Very concise, clear... amazing. Thank you!

[@DataMites]

You're so welcome!

[@trilokigupta821]

Great video, Excellent explanation..!

[@DataMites]

Thank you please watch Part video kzbin.info/www/bejne/emrcZ5eZrpWrgZo

[@bipulsinghkashyap8160]

Great video , explaination is very good .Thanks for this video

[@DataMites]

You are welcome!

[@hokapokas]

clear and great explaination i was looking for..Great Job.

[@DataMites]

Thank you!

[@alphavirginis260]

Excellent explanation! Thank you so much!!!

[@DataMites]

Please watch Part 2 here kzbin.info/www/bejne/emrcZ5eZrpWrgZo

[@farrasalyafi4370]

thanks, this is the best explanation so far

[@DataMites]

You're welcome!

[@probabilitycodingisfunis1]

thank you for the crystal clear explanation, loved it

[@DataMites]

Glad you enjoyed it!

[@akshayrajiyer]

Thank you very much! Excellent video with very clear explanation!

[@DataMites]

Thank you please watch another part video here kzbin.info/www/bejne/emrcZ5eZrpWrgZo

[@alamgirhasan5271]

Thank you so much. Nice Explanation. We want more videos like that

[@DataMites]

Keep watching

[@carmenrodriguez5020]

Excellent video and explaination, thank a lot!!

[@DataMites]

You are welcome!

[@sandile4764]

Thank you for a very good video.

[@DataMites]

Please watch Part 2 video kzbin.info/www/bejne/emrcZ5eZrpWrgZo

[@manikmahashabde2946]

Great explanation Sir. Thanks :)

[@DataMites]

Most welcome!

[@robertkulig5176]

Nicely done. Simple explanation, thank you!

[@DataMites]

Thank you. Please watch part video here kzbin.info/www/bejne/emrcZ5eZrpWrgZo

[@udayanrisingsun]

Nice Explanation ! specially by using Egs.. Also due to your slow speed , it is easy for beginners to understand.

[@DataMites]

Thank you

[@muhammadfuadbinhamzah1807]

thank you for detail explainantion

[@DataMites]

You are welcome

[@shubh13272anand]

What is difference between white noise and stationary data. As for white noise also mean and variance is constant and autocovariance does not depend on time. Also when mean=0 for white noise, then how average is best forecast for this series.

[@DataMites]

Yes, Shubham. You are right

[@janardhan9840202300]

Nice one !!! Do you have topics covered on SARIMA & using external varibles ?

[@DataMites]

Hi, We will post it soon

[@chandanjha3205]

Very nice.

[@DataMites]

Thank you

[@Adinasa2]

Where can I get this presentation PPT

[@DataMites]

Sorry. It is not possible

[@Adinasa2]

@@DataMites please share it. It will be very helpful for presentation

[@IAKhan-km4ph]

Brilliantly done. Sir, I am working on monthly and seasonally sea surface temperature data which forecasting methods would you like to suggest.

[@DataMites]

FBProphet will give better results. You can even explore LSTM for the same.

[@TheNishi42]

Nice,do you have any video on Multivariate Time Series also ?

[@DataMites]

Hi Romita Thakur, keep checking our channel, we will upload that soon.

[@TheNishi42]

@@DataMites no video uploaded on time series from so long on your channel hope to see more videos soon.

[@krishnakrmahto97]

you said for white noise data average is the best way we can talk about it, since time series won't be helpful. But you also told mean=0 for white noise data?

[@DataMites]

A series is called white noise if it is random in nature. let {𝜺1} denote such a series then it has zero mean [E(𝜺1)=0], has a constant variance [V(𝜺1)= 𝝈^2] and is an uncorrelated [E(𝜺t, 𝜺s) = 0] random variable. in other words, Time series that show no autocorrelation is called white noise.

[@LUMIGOCHA]

Excellent video and great explanation. I have a question: Best AIC is equall to lowest RMSE? If not, why not choose the modelo that give you the lowest error at the validation dataset? Good day for all.

[@DataMites]

"It is possible to have higher AIC even when you have a lower RMSE. In the given example, we have very less validation dataset compared to the training dataset, in this case, your result will be biased towards your validation set. For time series forecasting we usually take a large portion of data for training to capture the pattern in data more precisely, so it's better to go with error associated with the training set."

[@nithinmamidala]

please check the ppt .moving average you written a formula different and when you combine the AR MA formulas you wrote is different... please make a sense..

[@nanazeethiopia2892]

Nice .Thanks.

[@DataMites]

You are welcome!

[@javariailyas2429]

Is there any lecture about Matplotlib and SARIMA

[@DataMites]

we have a video tutorial on ARIMA, and matplot is used through the video lectures in the youtube instead of one dedicated video on it.

[@devyaninitturkar8662]

Thank you sir.. Can you please explain vector auto regression . If we data of different products and thier monthwise sale values for last few year ..then how to find result by using python

[@DataMites]

Sure. We will do it in upcoming videos

[@mehnazhazarika6611]

can anyone tell me what value we have to take for the constat in autoregressiv model ??

[@DataMites]

"Hi, thanks for reaching us out in your doubt. If you are talking about Auto-Regressive Integrated Moving Average movel, you need to find out the optimal value for p,d and q."

[@TheMunishk]

One thing i am still struggling is to how to calculate stock price using excel and apart from closing price on time series which other attributes I should include in time series. Example should I include market volatility or any other market benchmark. Can you suggest something?

[@edgarpanganiban9339]

Time series only includes the (time) and the variable you want to forecast or predict. In short, you dont need any other attributes, you rely only on the observation recorded through time. If you will include any other attributes or variable, then its not time series anymore...

[@amirdriv3r]

sound is not good man, change your mic....

[@mrmuranga]

content is good..but the mouse was really irritating....

[@TheMacsaucey]

Terrible presentation. Figure out a way to write clearly