# Unexpected negative output response to technology shock

**URL:** <https://forum.dynare.org/t/unexpected-negative-output-response-to-technology-shock/27445>\
**Category:** Stochastic simulations\
**Created:** [15 April 2025 12:44 UTC](https://forum.dynare.org/t/unexpected-negative-output-response-to-technology-shock/27445 "2025-04-15T12:44:19Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![Vicky](https://forum.dynare.org/letter_avatar_proxy/v4/letter/v/b19c9b/32.png) [@Vicky](https://forum.dynare.org/u/Vicky)\
**Post date:** [15 April 2025 12:44 UTC](https://forum.dynare.org/t/unexpected-negative-output-response-to-technology-shock/27445/1 "2025-04-15T12:44:20Z")

</div>

Dear community,

Hello. I am new to DSGE and Dynare, and I am working on a model that incorporates the financial accelerator, drawing from Christensen and Dib (2008) and CMR (2014).

I’ve encountered some strange results in the IRFs: for example, a positive technology shock (eta\_A) leads to a negative initial response of output (y).

 ![Figure 5 - Orthogonalized shock to eta_A figure 1](https://forum.dynare.org/uploads/default/original/3X/9/f/9f95f0bba1b8d9a25b3eddbb162a97ada1f59548.png)

I suspect this may be due to a timing mistake in the codes, but after carefully checking the equations, I haven’t been able to identify the problem. I have also consulted these related posts: [Production negatively react to technology shock](https://forum.dynare.org/t/production-negatively-react-to-technology-shock/20981), [Why the starting point of the technology shock to output is negative](https://forum.dynare.org/t/why-the-starting-point-of-the-technology-shock-to-output-is-negative/11813), and have simplified the model as much as possible, but the issue persists.

I would be very grateful for any suggestions on how to diagnose and correct this issue. I have attached my codes for computing the steady state and the .mod file:  
Steady state:  
[ss\_get.m](https://forum.dynare.org/uploads/short-url/d8L5thV5wyvrjy4NT6KrJhSkhNr.m) (6.5 KB)  
[ss\_get\_func4.m](https://forum.dynare.org/uploads/short-url/vRpqETrrFF1BYpHdHhkgXGlQT2f.m) (1.5 KB)  
Model:  
[model.mod](https://forum.dynare.org/uploads/short-url/6UY94jEyazRgitnh6TJitVQzGzS.mod) (13.8 KB)

Thank you very much for your help!

---

<div class="post-metadata">

**Author:** ![jpfeifer](https://forum.dynare.org/user_avatar/forum.dynare.org/jpfeifer/32/5044_2.png) [@jpfeifer](https://forum.dynare.org/u/jpfeifer)\
**Post date:** [16 April 2025 10:13 UTC](https://forum.dynare.org/t/unexpected-negative-output-response-to-technology-shock/27445/2 "2025-04-16T10:13:16Z")

</div>

Why is capital not predetermined in your model, e.g. in the production function?

---

<div class="post-metadata">

**Author:** ![Vicky](https://forum.dynare.org/letter_avatar_proxy/v4/letter/v/b19c9b/32.png) [@Vicky](https://forum.dynare.org/u/Vicky)\
**Post date:** [16 April 2025 12:25 UTC](https://forum.dynare.org/t/unexpected-negative-output-response-to-technology-shock/27445/3 "2025-04-16T12:25:34Z")

</div>

Hello Professor @jpfeifer , thanks for your reply. Capital is not predetermined in my model because in the reference papers, Christensen and Dib (2008) and CMR (2014), it seems that capital is not predetermined. I have also tried to have capital predetermined by changing `k` to `k(-1)` in the production function and the firm’s FOC of capital (code: [model1.mod](https://forum.dynare.org/uploads/short-url/ll6gV4EULfqk4EUKQcEqkTiDHJW.mod) (13.8 KB)  
), but then the IRFs do not change.

---

<div class="post-metadata">

**Author:** ![jpfeifer](https://forum.dynare.org/user_avatar/forum.dynare.org/jpfeifer/32/5044_2.png) [@jpfeifer](https://forum.dynare.org/u/jpfeifer)\
**Post date:** [16 April 2025 12:32 UTC](https://forum.dynare.org/t/unexpected-negative-output-response-to-technology-shock/27445/4 "2025-04-16T12:32:28Z")

</div>

Then that is likely your problem. The Christensen and Dib paper is not very consistent in outlining the proper timing.

---

<div class="post-metadata">

**Author:** ![Vicky](https://forum.dynare.org/letter_avatar_proxy/v4/letter/v/b19c9b/32.png) [@Vicky](https://forum.dynare.org/u/Vicky)\
**Post date:** [16 April 2025 13:25 UTC](https://forum.dynare.org/t/unexpected-negative-output-response-to-technology-shock/27445/5 "2025-04-16T13:25:29Z")

</div>

Thank you for your suggestion, Professor. I am aware of some issues in the timing of the Christensen and Dib paper. I will keep checking.