## Eviews 9

For technical support, tips and tricks, suggestions, or any other information regarding the EViews model object.

Moderators: EViews Gareth, EViews Moderator, EViews Chris

Kam_S
Posts: 2
Joined: Mon May 02, 2016 8:25 pm

### Eviews 9

I have installed Eviews 9 recently. I tried to solve a model using Eviews 9. I got the following message:Solve terminated - Unable to evaluate Jacobian at starting values. However, if I run the same model in Eviews 8 using the same options for model solution, I do not have any problems. Bugs in Eviews 9?

Eview 8 :
Model: DYN
Date: 05/03/16 Time: 16:16
Sample: 2015Q1 2037Q2
Solve Options:
Dynamic-Deterministic Simulation
Forward Solution
Solver: Broyden
Max iterations = 5000, Convergence = 1e-08

Parsing Analytic Jacobian:
81 derivatives kept, 4 derivatives discarded

Scenario: dynn
Solve begin 16:16:50
Pass 1 - 70 successful 16:16:51
Pass 71 - 145 successful 16:16:52
Pass 146 - 200 successful 16:16:52
Solve complete 16:16:52

Eview 9

Model: DYN
Date: 05/03/16 Time: 15:16
Sample: 2015Q1 2037Q2
Solve Options:
Dynamic-Deterministic Simulation
Forward Solution
Solver: Broyden
Max iterations = 5000, Convergence = 1e-08

Scenario: dynn
Solve begin 15:16:46
2015Q1 Block 1 - 13 eqns Solved (recursive block)
2015Q1 Block 2 - 3 eqns Convergence after 1 iterations
2015Q1 Block 3 - 1 eqns Solved (recursive block)
2015Q1 Block 4 - 3 eqns Convergence after 1 iterations
2015Q1 Block 5 - 5 eqns Solved (recursive block)
2015Q1 Block 6 - 4 eqns Convergence after 1 iterations
2015Q1 Block 7 - 10 eqns Solved (recursive block)
Solve terminated - Unable to evaluate Jacobian at starting values.

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 11811
Joined: Tue Sep 16, 2008 5:38 pm

### Re: Eviews 9

Hard to say without seeing the workfile.

Kam_S
Posts: 2
Joined: Mon May 02, 2016 8:25 pm

### Re: Eviews 9

Hi Gareth

Here is the workfile and the model is called dyn.
Attachments
model2.wf1

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 11811
Joined: Tue Sep 16, 2008 5:38 pm

### Re: Eviews 9

Thanks, we'll investigate.

I will note that it seems to solve under Gauss-Seidel, or with numerical derivatives.