I use the bitcoin price and other block chains data to do the Johansen Cointegration test. The result show that they have two Cointegration. Then, I construct the VECM, but I don't know how to interpret the results. What is the "ect2" means?
Dear Guanfu Chou,
You may first assess whether ect1vand ect2 are significant or not based on p values. Then you should look for their positive/ negative signs. Other test for model stability may also be carried out. Follow the prescribed process scrupulously.
Watching this video might help;
https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&cad=rja&uact=8&ved=2ahUKEwjg8Yfc5aL4AhVBOhoKHbFfCEAQwqsBegQIAhAB&url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3Dl-0DgBzdKLQ&usg=AOvVaw23Jn18FjCVarkT8NTGbgvl
That is econometrics without economics. Before you estimate something, you should know what you want to analyse and make clear, how you think (or economic theory says) relevant variables relate to each other. Where is this specification of your model? For this specification, you need no data (the specification should tell you, which data may be relevant), but before beginning to apply complicated methods, one should analyse simple relations between the data (via diagrams, correlations, OLS).
Why do you use error correction? How do you or the program (?) calculate ect1 and ect2? If I applied ec, I would explicitely formulate the relation.
Without knowing your variables, one can easily see, that your estimation is worthless: none of the coefficients is significant at any satisfying probability level.
- Badges
- Science topic
- Similar topics
- Vectorization