Hi,
Has someone programmed or found a program procedure for Ruey Tsay's non-linearity TAR-F test?
Thx,
Thomas
Program procedure for TAR-F test
Moderators: EViews Gareth, EViews Moderator, EViews Jason, EViews Matt
Re: Program procedure for TAR-F test
I'm particulary having trouble programming a routine for the arranged regressions. How can I rearrange a time series, y, so that I get time sequences related to a AR(p) process for the smallest y(j), then the second smallest y(j) and so on, where j is the index that orders the series from smallest observation to largest observation.
Logically, the program would have to search for the smallest y(j), y(1), in the sample and then order the time-congruent AR(p) sequence for that y(1) in a row vector.
Is there a command that searches for smallest, second smallest, so on, in a time series and then "cuts out" the adjacent prior p observations into a vector?
Or alternatively, delivers the corresponding date. With that date I can then search for the adjacent prior p observations and form a vector.
Still searching,
T
Logically, the program would have to search for the smallest y(j), y(1), in the sample and then order the time-congruent AR(p) sequence for that y(1) in a row vector.
Is there a command that searches for smallest, second smallest, so on, in a time series and then "cuts out" the adjacent prior p observations into a vector?
Or alternatively, delivers the corresponding date. With that date I can then search for the adjacent prior p observations and form a vector.
Still searching,
T
Who is online
Users browsing this forum: No registered users and 2 guests