# Steady State of log-linearized model

**URL:** <https://forum.dynare.org/t/steady-state-of-log-linearized-model/11349>\
**Category:** Steady state\
**Created:** [8 February 2018 11:16 UTC](https://forum.dynare.org/t/steady-state-of-log-linearized-model/11349 "2018-02-08T11:16:15Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![marlenemaya](https://forum.dynare.org/letter_avatar_proxy/v4/letter/m/3bc359/32.png) [@marlenemaya](https://forum.dynare.org/u/marlenemaya)\
**Post date:** [8 February 2018 11:16 UTC](https://forum.dynare.org/t/steady-state-of-log-linearized-model/11349/1 "2018-02-08T11:16:15Z")

</div>

Good morning to everyone!  
I’m trying to replicate the model in _Rubart, Heterogeneous labor, labor market frictions and employment effects of technological change_  
the model should be solved in deviation from the steady state.  
dynare is unable to find a steady state. What is wrong? I attach here the code  
the problem (one of the problems) is that anytime I’ve 1-exp(var) at the denominator in an equation this gives me inf for the steady state, but i suppose this is the correct way to log-linearize the model.

Thank you so much!  
[treloglin.mod](https://forum.dynare.org/uploads/default/original/2X/1/145dd8ad2e293fbe707643cc4096352510c52e4e.mod) (4.5 KB)

---

<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:** [11 February 2018 13:04 UTC](https://forum.dynare.org/t/steady-state-of-log-linearized-model/11349/2 "2018-02-11T13:04:05Z")

</div>

No, this is not a proper way of loglinearization. It seems you are trying to do an exp()-substitution, but in this case, the steady state values are not 0. You may want to consult [Linearisation vs Log-linearisation](https://forum.dynare.org/t/linearisation-vs-log-linearisation/10535/3) and [Question about understanding irfs in dynare](https://forum.dynare.org/t/question-about-understanding-irfs-in-dynare/10622/4) for your options.

---

<div class="post-metadata">

**Author:** ![marlenemaya](https://forum.dynare.org/letter_avatar_proxy/v4/letter/m/3bc359/32.png) [@marlenemaya](https://forum.dynare.org/u/marlenemaya)\
**Post date:** [13 February 2018 23:14 UTC](https://forum.dynare.org/t/steady-state-of-log-linearized-model/11349/3 "2018-02-13T23:14:42Z")

</div>

Thank you! I think I understood what was wrong…but the problem is that I still don’t get 0 residuals in all my equations. I don’t really understand where the misteke is. Thank you[treloglin.mod](https://forum.dynare.org/uploads/default/original/2X/d/d766909f5acafcd739bd95940c63e4db62f1c0ab.mod) (5.8 KB)

---

<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:** [14 February 2018 07:59 UTC](https://forum.dynare.org/t/steady-state-of-log-linearized-model/11349/4 "2018-02-14T07:59:48Z")

</div>

For example

```auto
exp(log(p_s)) = exp(log(m_s))/((exp(log(s_s))*(1-exp(log(h_s))))); 

```

is not a linear(ized) equation

---

<div class="post-metadata">

**Author:** ![marlenemaya](https://forum.dynare.org/letter_avatar_proxy/v4/letter/m/3bc359/32.png) [@marlenemaya](https://forum.dynare.org/u/marlenemaya)\
**Post date:** [14 February 2018 08:33 UTC](https://forum.dynare.org/t/steady-state-of-log-linearized-model/11349/5 "2018-02-14T08:33:17Z")

</div>

Thank you! So do you think I should manually linearize those equations or avoid to apply log linearization?  
Thank you!