The RATS Software Forum

Dear Tom Doan,

I am using tests for (periodically repeating) mildly explosive behavior (rho > 1 but close to 1). Phillips et al (2011, 2012,2015) propose several ADF-type tests (recursive, rolling window) and a MATLAB / GAUSS code is available here: https://sites.google.com/site/shupingshi/home/research . An Eviews Add-in also exists. Since I want to keep my reproduction files in one single program I ask for a replication code in RATS.


Reference:

Phillips, Peter C.B., Shu-Ping Shi and Yun Yu (2015): Testing for Multiple Bubbles: Historical Episodes of Exuberance and Collapse in the S&P 500, International Economic Review, 2015, 56 (4 ): 1043-1078

This does the two main tests. This uses RLS to keep the computational burden down. With the 2000 replications, the calculation of the moving critical values would take about 7x10^9 regressions if done as it's programmed in Gauss or Matlab. (I'm not sure why the SADF result doesn't quite match, though the timing on the reported GSADF range is probably due to the common error of not adjusting the date for the last time period in differencing when using a math package).

Hi

I just ran this and got this error message:

## SX11. Identifier SADF is Not Recognizable. Incorrect Option Field or Parameter Order?
>>>># sadf<<<<
Is it possible you meant
BSADF
If the name isn't mistyped, it's possible that you have a poorly formatted instruction
Common errors are
* a space before the ( in an option field
* a missing space before = in a SET or FRML
* a missing $ at the end of a long line which continues to the next

I uploaded a new version. Try that.

Hi

Just tried it. Works. Graph like fig 8.

Thanks a lot. Seasonal greetings, Ulrich Fritsche