Hi,
when estimating VAR with endogenous and exogenous variables the following error pops out.
The same is with CUSUM of squares and recursive residuals. At the end the graph is done but with this error.
i think that something is wrong when exogenous variables are present.
Manny
Bad dimension of restriction matrix S_B (II)
Bad dimension of restriction matrix S_B (II)
Bad dimension of restriction matrix S_B (II)
Bad dimension of restriction matrix S_B (II)
Bad dimension of restriction matrix S_B (II)
6491492 ERROR [Thread-92] com.jstatcom.engine.PCall - failed call: VAR CUSUM
java.lang.RuntimeException: procedure: "stab_VARcusum_wrapper"
arguments:
arg_1: com.jstatcom.model.JSCNArray@18ec029 [name=varBuffer,type=NARRAY,value=1092x1,[0,0]=NaN]
arg_2: com.jstatcom.model.JSCInt@e9c592 [name=startPoint,type=INT,value=23]
arg_3: com.jstatcom.model.JSCNumber@1a9fcea [name=startDate,type=NUMBER,value=1980.4]
arg_4: com.jstatcom.model.JSCInt@11b99c4 [name=period,type=INT,value=4]
arg_5: com.jstatcom.model.JSCNumber@10c29fe [name=signLevel,type=NUMBER,value=0.99]
arg_6: com.jstatcom.model.JSCInt@1991ba7 [name=isGraph,type=INT,value=1]
arg_7: com.jstatcom.model.JSCInt@e674d6 [name=method,type=INT,value=0]
Bad dimension of restriction matrix S_B (II)
Bad dimension of restriction matrix S_B (II)
Bad dimension of restriction matrix S_B (II)
Bad dimension of restriction matrix S_B (II)
Bad dimension of restriction matrix S_B (II)
at com.jstatcom.engine.grte.GRTEngine.call(Unknown Source)
at de.jmulti.proc.VARCusumCall.runCode(Unknown Source)
at com.jstatcom.engine.PCall.run(Unknown Source)
at EDU.oswego.cs.dl.util.concurrent.QueuedExecutor$RunLoop.run(Unknown Source)
at java.lang.Thread.run(Unknown Source)
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Also,
when re-estimateing subset VAR (with intercept and trend included) with new number of lags, coefficients in subset VAR pannel are reseted but intercept and trend coefficients NOT!
Manny
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi,
when estimating VAR with endogenous and exogenous variables the following error pops out.
The same is with CUSUM of squares and recursive residuals. At the end the graph is done but with this error.
i think that something is wrong when exogenous variables are present.
Manny
Bad dimension of restriction matrix S_B (II)
Bad dimension of restriction matrix S_B (II)
Bad dimension of restriction matrix S_B (II)
Bad dimension of restriction matrix S_B (II)
Bad dimension of restriction matrix S_B (II)
6491492 ERROR [Thread-92] com.jstatcom.engine.PCall - failed call: VAR CUSUM
java.lang.RuntimeException: procedure: "stab_VARcusum_wrapper"
arguments:
arg_1: com.jstatcom.model.JSCNArray@18ec029 [name=varBuffer,type=NARRAY,value=1092x1,[0,0]=NaN]
arg_2: com.jstatcom.model.JSCInt@e9c592 [name=startPoint,type=INT,value=23]
arg_3: com.jstatcom.model.JSCNumber@1a9fcea [name=startDate,type=NUMBER,value=1980.4]
arg_4: com.jstatcom.model.JSCInt@11b99c4 [name=period,type=INT,value=4]
arg_5: com.jstatcom.model.JSCNumber@10c29fe [name=signLevel,type=NUMBER,value=0.99]
arg_6: com.jstatcom.model.JSCInt@1991ba7 [name=isGraph,type=INT,value=1]
arg_7: com.jstatcom.model.JSCInt@e674d6 [name=method,type=INT,value=0]
Bad dimension of restriction matrix S_B (II)
Bad dimension of restriction matrix S_B (II)
Bad dimension of restriction matrix S_B (II)
Bad dimension of restriction matrix S_B (II)
Bad dimension of restriction matrix S_B (II)
at com.jstatcom.engine.grte.GRTEngine.call(Unknown Source)
at de.jmulti.proc.VARCusumCall.runCode(Unknown Source)
at com.jstatcom.engine.PCall.run(Unknown Source)
at EDU.oswego.cs.dl.util.concurrent.QueuedExecutor$RunLoop.run(Unknown Source)
at java.lang.Thread.run(Unknown Source)
Also,
when re-estimateing subset VAR (with intercept and trend included) with new number of lags, coefficients in subset VAR pannel are reseted but intercept and trend coefficients NOT!
Manny
JMulTi still has bugs...
And the development seems to have stopped
What's for sure, is that wonderful forum is dying, I regret to say.
Yes, the biggest error was made when shortcut for this forum was replaced from the homepage...
Manny,
I agree with you, the developers need to hire a new grad student to keep JMulTi running smoothly.
Ken
as always, money matters. I cant believe Lutkephol is letting this happening