e3485e8929
Changes around normalizing Jacobians:
...
-) never use normalization for SVD and brute force covariance checks;
-) Jacobians are normalized for checking the rank condition: this should allow for more uniformity in results across models.
2011-09-01 11:18:24 +02:00
d83c5326f6
check that theoretical moments are independent when evaluating simulated moments uncertainty
2011-06-24 10:58:04 +02:00
756594fdc4
changed default ar=1 and perform automatic increase of ar if this is not sufficient
2011-06-15 23:47:27 +02:00
e5c3decbc1
trap for too many parameters to be estimated w.r.t. affected moments.
2011-06-06 16:36:39 +02:00
2c977073d2
Bug fix: thanks to Johannes Pfeifer
2011-06-06 15:21:59 +02:00
58dc9557d9
Fixes around point-estimate of sensitivity and identification strength
2011-05-30 14:36:48 +02:00
a45533caca
Added SVD for J matrix as output identification info.
2011-05-10 10:11:25 +02:00
cbfc017356
unix end of lines for new functions!
2011-05-02 11:32:26 +02:00
180a3bad9c
New wrapper for identification analysis, given the parameter set provides results.
...
Allows more flexible coding and checks for particular parameter sets.
2011-05-02 11:03:18 +02:00
Marco Ratto