When explaining Time Series modeling on Python, anyone would assume and directly say this pd.read_csv standard function. But he takes time in explaining the attributes of this function even for a basic beginner to understand which is something worthy. :)
Hi, but what if we dont have data for all the months like there is for 10 months but 2 months the sales wasn't done for the product so it wasnt added? Like if I have monthly wise data instead of daily what can I do? Like I have date like jan 2019 and so. If I convert to datetime in python it shows 01-01-2019 for jan and then all months start with 01 for the following year.
Hi!, thanks a lot for your videos, they're quite helpful! Do you have any other textbook recommendations for learning timeseries forecasting? The ones you've mentioned are practical/more python centric, I'm interested in mathematical/theoretical textbooks. Thanks!
Say my dataset is not stationary and from PACF it's observed that upto 4th lag shows good correlation. If it was ARIMA I could give these as parameters of ARIMA Order model function, but what to do in AR model is there a similar way to pass as parameter. Or do you just simply difference by using shift method by 4 and use that set of data as values to be fed into the train/test set?
Thank you for this video, very good explanation! One question: If I decide to include only 5 lags in my model and I want to predict the temperature for the next 7 days, does this mean that the predictions for day 6 and day 7 are only based on the models' predictions for the days 1-5 and on no observed value? Can this result in some sort of "echo chamber" effect?
How we can predict deforestation by studying the NDVI index from time series from the Google Earth engine? prediction for next years that we have no data on that, we have data only from the past until now!
Actually I am trying to forecast 3 sensor values for the next 500 time slots using LSTM,but it shows lot of noise...how can I proceed or should I change my approach?
Oh my God! The best videos ever on Time Series Forecasting! Love all of them. Thanks a million!
Bro, you just saved 200 students from failing the exam💯
Thank you for making this series.
When explaining Time Series modeling on Python, anyone would assume and directly say this pd.read_csv standard function. But he takes time in explaining the attributes of this function even for a basic beginner to understand which is something worthy. :)
can u provide the github link for the same?
thanks!
(need it urgently) :)
This was an Extremely helpful video with clear explanation. Thanks a lot man
Bhagwaan bless you with great health bro. 🙏
This is explained very well
you are literally AMAZING man
Haha, thanks!
Thanks a lot !!!!!!!!!!!!!!!!! Very helpful and clear explanation!
Hi, I am not finding the dataset used here. Please guide me where to find it.
where is the google colab notebook?
at 3:07 on the plot we see in the y axis values from 0-25 i guess which are temp values but how exactly did you put them there? i dont understand it
Amazing videos! Thank you!
higher values of correlation means higher correlation agreed, but higher negative values also means the same. Only values around 0 are not correlated.
Great content sir
If you have provided the resources in Github that will help so many
Brilliantly explained, but it would be better if you used AIC to select the lag..
Thanks a ton. Explaination is superb.
Could you please explain the significance of dynamic being set to False inside the model.predict function
Hi, but what if we dont have data for all the months like there is for 10 months but 2 months the sales wasn't done for the product so it wasnt added?
Like if I have monthly wise data instead of daily what can I do? Like I have date like jan 2019 and so. If I convert to datetime in python it shows 01-01-2019 for jan and then all months start with 01 for the following year.
What is autolag="AIC" that you used in ADFULLER TEST?
absolutely amazing!
Hi!, thanks a lot for your videos, they're quite helpful!
Do you have any other textbook recommendations for learning timeseries forecasting? The ones you've mentioned are practical/more python centric, I'm interested in mathematical/theoretical textbooks. Thanks!
Correction at 4:50, p value should be
Hi, where can I get the csv file of this, which you have used in this video?
How can i get this Data Set ?
Hie Hebber, I got the ACF plots positive (Constant ). what should i do?
Great guide! Thank you!
Say my dataset is not stationary and from PACF it's observed that upto 4th lag shows good correlation. If it was ARIMA I could give these as parameters of ARIMA Order model function, but what to do in AR model is there a similar way to pass as parameter. Or do you just simply difference by using shift method by 4 and use that set of data as values to be fed into the train/test set?
is there a video about ARMA MODELS??? kindly reply
very nice, keep posting...waiting for deployment videos of forecasting model on azzure
Thanks, will try to make a video on ir
Thank you so much for this video. I was wondering how to apply auto reg on groups in datagrams, any suggestions will help alot
Can you help me with the persistence model? in solar energy forecasting. i have requested in linkedin.
Very good. You would make an excellent teacher. Get a PhD in USA and join teaching at a University.
how i get this dataset file?
Thank you for this video, very good explanation! One question: If I decide to include only 5 lags in my model and I want to predict the temperature for the next 7 days, does this mean that the predictions for day 6 and day 7 are only based on the models' predictions for the days 1-5 and on no observed value? Can this result in some sort of "echo chamber" effect?
what to do if the probability is high?
can you make a video without using inbuilt function
How we can predict deforestation by studying the NDVI index from time series from the Google Earth engine? prediction for next years that we have no data on that, we have data only from the past until now!
Please let us know what value should be taken for lag?
Incase if there is non-stationarity in the data then, where to write the order of differencing in code? at the time of fit.
dear please help me how to add CSV file in google colab?
Such a cool video bro...u r awesome :)
You have the research collab, why not share it... its 2023 dude
Actually I am trying to forecast 3 sensor values for the next 500 time slots using LSTM,but it shows lot of noise...how can I proceed or should I change my approach?
Start with smoothing data
simple awesome
4:40 i think you mean greater than 0.05, not 0.5
Bro, you are amazing 👌👏👏👏👏👏👏👏👏👏👏👏👏👏👏👏👏👏👏👏👏
Thanks a lot!
Nachiketa, I think so p-value should be 0.05 and not 0.5. let me correct if I am wrong
yaa, it is 0.05
Can u please give us code of this tutorial
Pls share the data set
thanx
can u provide github link?
Why lags=10?
Dude, you saying p-value should be less than 0.5 ..... but you mean 0.05 9:06
👍
i just cannot under your accent, sorry
The predictions are way off