Econometrics: why co-integration is needed - page 16

 
Farnsworth:

to faa


How will you co-integrate? Let me guess - assign it to EW, and what he will do is pull it down by the ears within almost any model and you will get a fake stationarity.

Why fake? Millions of users have the package and no one has said "fake". And a person who has not seen the package is fake?

 
faa1947:
Stationarity is checked by the unit root test. Existing subtleties that occur due to ACF are solved within the test or by the choice of test type (I have several of them at my disposal). I see no reason not to use existing achievements and start repeating what was done 20 years ago e.g. by Hamilton
Stationarity is tested in a bunch of ways this is one of them. Just do what is asked. Is it difficult?
 
faa1947:

How will you co-integrate? Let me guess - assign it to EW and what he will do is pull it down by the ears within almost any model and you will get a fake stationarity.

Why a fake one? Millions of users of the package and no one said "fake". And the person who hasn't seen the package is fake?

:))))))

Are you reminded of your post where after another failure you call this package the lowest, most primitive etc. It's not the package, it's the layer between the monitor and the chair. Sort of a joke.

OK, you don't want to do that. It's your problem and it's just beginning :)

 
tara:

Sergei, did your father or mother teach you to be rude?

Sneaking around? Alexei, what makes you think I'm being rude? I'm trying to give advice to the best of my ability, that's all. I responded to the ignoring and proffessor's kicking and trying to explain what he himself does not understand with a good joke. I pointed out where the problem is in this case. You think I'm not self-critical? Wrong and I am sometimes such a "layer", it's okay, as long as there is searching and comprehension.

But still you have a keen sense of justice.

 
Farnsworth:
Stability is checked in a number of ways this is one of them. Just do what is asked. Is it difficult?

The package has 39 sections, each describing some econometric methods + three appendices. Each section has a reference list of 10 to 20 titles, where the algorithms implemented in the section are described.

If we talk about unit root test, I have 6 types of test, which is performed for three levels of integration with or without trend bias included in the test, with 7 criteria for choosing the number of lags (people also understood about ACF). Why do you keep lecturing me? If there is any constructive feedback, go ahead. But don't try to make me develop new methods of determining stationarity. They have been developed a long time ago - you just have to take them and do meaningful things, e.g. justify the credibility of the tester results based on the stationarity of the residual. Or anything else.

 
Farnsworth:

a sense of justice.


feeling. I can...
 
faa1947:

The package has 39 sections, each describing some econometric methods + three appendices. Each section has a reference list of 10 to 20 titles, where the algorithms implemented in the section are described.

If we talk about unit root test, I have 6 types of test, which is performed for three levels of integration with or without trend bias, with 7 lag selection criteria (people also understood about ACF). Why do you keep lecturing me? If there is any constructive feedback, go ahead. But don't try to make me develop new methods of determining stationarity. They have been developed a long time ago - you just need to take them and do meaningful things, e.g. justify the credibility of the tester results based on the stationarity of the residual. Or anything else.

Yes, I use the bag from time to time, I have a very good idea of what it is, it's not about the bag.

Faa, don't faint, you don't want to. Once again, I do not need it, it's your problem.

When are you going to start giving constructive feedback? The market will teach you, in this sense I had enough of your last topic and the model with 3 coefficients. What a lot of effort it took to convince you that this is bullshit.

PS: and stop hiding behind authority (you are not able to do it well), write your thoughts. And I strongly recommend - also look at the data.

 
Don't dare test stationarity on yourself. It's bad luck.
 

To faa

In the end. There is no ACF stationarity in your series and there cannot be. You will not be able to predict it, it will have the same effect, 50/50.

Have fun.

 
paukas:
Don't dare test stationarity on yourself. It's bad luck.
Yeah, and if it's scattered ....