Login
Login

Forum Replies Created

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
    Posts
  • in reply to: CointegrationMLE #2039
    tchristell
    Member

    Thanks Ryu,

    In that case, I should get the same regression coefficients as if I had ran the regression on another software package like R, Matlab, Eviews etc. Curiously though, I am getting different numbers. Does Shuanshu ‘clean’ the data before running the regression, for example, getting rid of outliers above a certain threshold. Or does it run a regression on the pure data input to the program.

    I am assuming OLS regression here.

    dj

    in reply to: CointegrationMLE #2037
    tchristell
    Member

    Hi Kevin,

    I’m am in the same group as Ivan, exploring cointegration with algoquant. Correct me if I am wrong for the following:

    I assume that Algoquant:

    (1) runs a regression between Y and X (assume 2 time series data)
    (2) puts the residuals through the Johansen test and outputs the test statistic values in getStats()
    (3) outputs the regression coefficients as well

    dj

Viewing 2 posts - 1 through 2 (of 2 total)