Ignore:
Timestamp:
03/19/12 20:08:15 (2 years ago)
Author:
Lan Zagar <lan.zagar@…>
Branch:
default
Message:

Removed second computation of matrix inverse.

File:
1 edited

Legend:

Unmodified
Added
Removed
  • Orange/regression/linear.py

    r10575 r10576  
    295295        sigma_square = sse / (n - m - 1) 
    296296        # standard error of the regression estimator, t-scores and p-values 
    297         std_error = sqrt(sigma_square * pinv(dot(X.T, X)).diagonal()) 
     297        std_error = sqrt(sigma_square * invcov.diagonal()) 
    298298        t_scores = coefficients / std_error 
    299299        p_vals = [stats.betai(df * 0.5, 0.5, df / (df + t * t)) 
Note: See TracChangeset for help on using the changeset viewer.