TAR model & asymmetric ECM

[ecomary]

Hello
I want to estimate TAR & M-TAR model and do asymmetric ECM. I have added tarcoint in my EViews. But I don't know how I must do asymmetric ECM by tarcoint. What must I write as endogenous and exgenous variables? And what steps must I do for doing esymmetric ECM?
and some other questions, when I want to estimate a TAR model by tarcoint, Must I only write my variables in endogenous and exgenous boxes and press OK? and are all my variabls my endogenous variables and are only c(intercept) my endogenous variables? and mustn't I do any steps for estimating a tar model for example like ADF test?
Pleas help me. I really need your replies. Thank you very much.
Best regards
Mary

[trubador]

Please read the help file before running the add-in (e.g. C:\Users\...\Documents\EViews Addins\tarcoint\tarcoint.pdf). And you will probably better off referring to the original study of Enders and Siklos (2001), which is cited in the References section of the help file and in the bottom right corner of the add-in's window.

[ecomary]

thank you so much dear trubador.

[ecomary]

dear trubador, I read that pdf in help of tarcoint, but I still have some questions and I am really sorry that I am bothering you again: I don't still know what I must write in endogenous and exogenous boxes when I want to estimate asymmetric ECM model. And what steps must I do for doing it by tarcoint, when I have 2 variables.
I know the theory of asymmetric ECM and I know how estimate symmetric ECM by EViews, but I don't know how I must estimate asymmetric ECM by tarcoint in EViews.
I really thank you very much for your replies, and if it is possible for you reply me again.
best regards
Mary

[trubador]

Those two variables are your endogenous variables and therefore you should enter them into the first box as seperate series (e.g. y1 y2) or as a group (e.g. yvar). If you have independent variables, which might have an impact on these dependent variables then you should enter them into the second box, again as seperate series (e.g. x1 x2) or as a group (e.g. xvar). You follow the same rules as you would do in estimating a symmetric ECM with EViews. There is nothing specific going on here in that respect. And if you are familiar with theory behind this model, then other options should be pretty self-explanatory.

[ecomary]

Thank you very much for all your replies.

[anne92]

I'm a student and a new users of Eviews. I want to test TAR and MTAR model to test cointegrate between goldprice( ln goldprice) and CPI ( ln CPI) . I've read the paper of Enders and Siklos and have done as the instruction . However, I have 2 question is that.
1. Endogenous I must fill is series ln gold price and ln CPI. Is that right?
2. As results I recieved a table with only threshold value , t-Max , F-phi . How about the p-value ? Is there p-value in the result table ? Have I done some error so I recieved not p-value?
I'm looking to hearing from you.

Best regard.

[trubador]

The answer to your first question might have become obvious after reading the paper. As for the second question, you can use the related tables in the original paper for hypothesis testing. The add-in offers a Monte Carlo study to obtain critical values, of which the details are outlined in the original study. As emphasized in the add-in's help file: "Each model specification/combination requires a separate Monte Carlo study to test the null hypothesis of no cointegration against the alternative of cointegration with threshold. If you wish to experiment for critical values, simply tick the related checkbox and supply the number of simulations along with a significance level. [...]"

If the test statistics are significant at the alpha level you supplied, then the related result will be highlighted.

[nishantvats12]

Hi,

I am currently working on TAR and M-TAR model. I have following issues, it would be really helpful if someone could guide me on that:
1. Could anyone please guide me how to get the exact value of AIC,SIC and RMSE for the model generated by the tarcoint package of EViews.
2. When I use more than 2 variables in the endogenous list (and also when as a group) the add in shows an error saying Vtemp not defined. There is no such thing mentioned in the help file of tarcoint. Could anyone please tell me why the problem is occurring and how it can be solved.
3. After estimating the asymmetric equation of the error term how does one go about estimating the actual error correction system (see equation 20. 21) in the original Enders Silos Paper https://virtualtour.wlu.ca/documents/14 ... s_jbes.pdf

Thanks in advance.

[nishantvats12]

Anyone with any help?

[esida]

i am working on the asymmetry of interest rate pass-through in the case of Albania adopting the methodology outlined in Enders and Siklos (2001). There is one thing i do not clearly understand. Why is delta1=delta2=0.6 (even in the tarcoint add-in such value is taken as default). i read the paper carefully but i still cant understand why they take such values.
I really would appreciate your help.

[nishantvats12]

You can change the value of delta as per your need and justification. The tarcoint package gives that autonomy to the user.

[msvanidze]

Hello,
Is there anybody to help me interpret TAR coefficients? Namely: "Above threshold", "Below threshold" and coefficients on differenced residuals. I am familiar with Asymmetric VECM and I know how to interpret its coefficients but there I got confused :(

[KOFIBLAY]

Hello
I want to estimate TAR & M-TAR model and do asymmetric ECM. I have added tarcoint in my EViews. But I don't know how I must do the TAR, M-TAR and asymmetric ECM by tarcoint. What must I write as endogenous and exgenous variables? And what steps must I do for doing esymmetric ECM?
and the other questions, when I want to estimate a TAR model by tarcoint, if I feed the series it pop-out that EQ1 is not defined. Please can anyone help me to overcome this problem. My data series is prices of a homogeneous product from five different markets and want to test the long-run relationship (cointegration)
Pleas help me. I really need your replies. Thank you very much.
Best regards

[trubador]

1) Carefully read the original paper: Enders and Siklos (2011)
2) See the help document for details before running the add-in.
3) Search the forum for similar discussions: (e.g. http://forums.eviews.com/viewtopic.php?f=23&t=5604)