Causal Impact Analysis in Time Series using R
HTML-код
- Опубликовано: 10 окт 2024
- This video goes through an example of Causal Impact Analysis for time series econometrics using the CausalImpact Package in R.
Created by Justin S. Eloriaga
Codes and Dataset here:drive.google.c...
Thanks a lot for this video. I have no use for it at the moment but I use R for social sciences and it’s always good to know more functions for R for future purposes. You do an excellent job of showing your steps and explain very clearly the results. Subscribed to your channel because of your knowledge on how you utilize R and explanations in other videos that are understandable.
All your videos are quite lucidly explained. Thank you for sharing your knowledge.
Looking forward to more videos on time series econometrics.
Thank you Justin for this informative video on Causal Impact Analysis. Could also demonstrate how to enter annual time series eg" 2007, 2008, 2009 etc from CVS file.
Just change frequency = 1 in the ts() command
TY so much for this video. Do you have resources for how to choose covariates?
Thank you so much sir keep posting models in R...very interesting. Sir, from where you are
Hi. Thank you for sharing such a informative video. But could not find the codes in the google drive link you shared. Can you please share the codes by creating a folder named "Codes" for this video content?
What kind of regression does the CausalImpact() function implement?
Hi. This is very helpful. Quick and straight to the point. May I just ask, is there a minimum number of data points required for this analysis both in the pre and post intervention periods? Would this work with a monthly time series with 12 months pre intervention and 3 months post?
Top content
What's a good method of choosing covariants?
Awesome video, explained very well, any idea if this can be done in python?
Hello! Thank you so much for the Video, this helps me very much for my masterthesis :) How will the time.points command for quarterly be? my Dataset starts with 1994 Q1 and ends with 2019 Q4. I have already converted my data from daily/monthly to quarterly data.
the original data does not include weekend dates like 29/02/2020 or 01/03/2020. However the "data" does. and the 44 days for the ts start from 27/02 to 10/4 rather than 28/04. Would that be a trouble? Maybe just use:
df$date
how can you make one with an unweighted average? what is the code for r
how does the model decide which one is the response variable (in this case : ibcl) ?
The first variable ordered in the cbind()
I keep getting data is not numeric. What could be the problem?
You may have N/As in the data or the date may not be formatted correctly. Kindly check out the lubridate commands.
@@JustinEloriaga I have formatted the dates according to your dataset but it's still giving me the same problem
where's the next session?