List view
- No due date•0/1 issues closed
- No due date•0/1 issues closed
We need this plugin to be able to get an Oauth2 token to read and write measurements to the QM database
No due dateNew answer to "What's the best R package for predicting causality between two streams of time-series data?" Prashanth Sriram, Data Scientist at Sabre Airline Solut... In my opinion, the best way to test this is using the Granger Causality Test. Here's the Wikipedia link to get you started off on understanding this test: Granger causality . I suggest you read some articles or textbooks on Econometrics involving multiple time series analysis to get a basic understanding of the concepts involved, such as Cointegration: Cointegration. Two implementations of the Granger causality test can be found in the open source package 'vars' in the R programming language. However, there are some limitations to when you can apply these tests, and there are some variations of the test that can get around them. A good discussion of this can be found in this article by Prof. Dave Giles: Page on davegiles.blogspot.ca . It is worth noting that he points out a few errors on the Wikipedia page on the Granger causality test I linked above. An R code implementation of the Toda-Yamamoto procedure for the Granger causality test as explained in Prof. Giles' page can be found in this article by Christoph Pfeiffer: Toda-Yamamoto implementation in 'R' Good luck.
No due date