# Perfect foresight estimation with histval

**URL:** <https://forum.dynare.org/t/perfect-foresight-estimation-with-histval/28317>\
**Category:** Perfect foresight simulations\
**Created:** [23 March 2026 21:54 UTC](https://forum.dynare.org/t/perfect-foresight-estimation-with-histval/28317 "2026-03-23T21:54:48Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![ani.98](https://forum.dynare.org/letter_avatar_proxy/v4/letter/a/db5fbb/32.png) [@ani.98](https://forum.dynare.org/u/ani.98)\
**Post date:** [23 March 2026 21:54 UTC](https://forum.dynare.org/t/perfect-foresight-estimation-with-histval/28317/1 "2026-03-23T21:54:48Z")

</div>

Hi everyone, I am trying to create simulations with perfect foresight by giving initial values to the state variables in histval. Unfortunately when I provide values to some state variable either I get homotopy issue or unreasonable dynamics. the `model_info` command shows the state variables, some of which are price levels (P\_NTSPI for example) and relative prices (RP\_flex), I wonder if I should give every state variable an initial value or not? the model is heavy and i don’t want to complicate it more. Here are the state variables:  
`PIE_FLEX(-1)`  
`PIE4_NTSPI(-1)`  
`PIE_PRT(-1)`  
`P_NTSPI(-1)`  
`P(-1)`  
`P_REG(-1)`  
`P_FLEX(-1)`  
`P_food_imp(-1)`  
`P_oil_imp(-1)`  
`P_PRT(-1)`  
`P_F(-1)`  
`RP_REG_t(-1)`  
`gRP_REG_t(-1)`  
`RP_REG_g(-1)`  
`RP_FLEX_t(-1)`  
`gRP_FLEX_t(-1)`  
`RP_FLEX_g(-1)`  
`qoil_t(-1)`  
`d_qoil_t(-1)`  
`qoil_g(-1)`  
`qfood_t(-1)`  
`d_qfood_t(-1)`  
`qfood_g(-1)`  
`EPS_PIE_NTSPI(-1)`  
`P_HEADLINE(-1)`  
`C(-1)`  
`EPS_PIE_NTSPI_EXP(-1)`  
`EPS_C(-1)`  
`PIE_STAR(-1)`  
`Y(-1)`  
`EPS_Y(-1)`  
`CUMSUMY(-1)`  
`DL_GDP_BAR(-1)`  
`G_GDP_BAR(-1)`  
`UNR_GAP(-1)`  
`G_UNR(-1)`  
`EPS_UNR_BAR(-1)`  
`RS(-1)`  
`TERM6M(-1)`  
`TERM1Y(-1)`  
`TERM2Y(-1)`  
`TERM3Y(-1)`  
`TERM5Y(-1)`  
`TERM10Y(-1)`  
`RR_BAR(-1)`  
`EFF_RR_BAR(-1)`  
`sigma(-1)`  
`EPS_RS(-1)`  
`P_GAP(-1)`  
`PREM(-1)`  
`RR_US(-1)`  
`RS_US(-1)`  
`PREM_BAR(-1)`  
`FX_PREM(-1)`  
`PIE_US(-1)`  
`P_US(-1)`  
`Z(-1)`  
`Z_t(-1)`  
`dZ_t(-1)`  
`Z_US(-1)`  
`Z_US_g(-1)`  
`Z_US_t(-1)`  
`Z_PRT(-1)`  
`Z_PRT_g(-1)`  
`Z_PRT_t(-1)`  
`dZ_PRT_t(-1)`  
`S(-1)`  
`S_PRT(-1)`

```auto
P_NTSPI(-2) (original expression P_NTSPI(-1)) 
P_NTSPI(-4) (original expression AUX_ENDO_LAG_14_1(-1)) 
P_NTSPI(-6) (original expression AUX_ENDO_LAG_14_2(-1)) 
P_FLEX(-2) (original expression P_FLEX(-1)) 
P_FLEX(-4) (original expression AUX_ENDO_LAG_17_1(-1)) 
P_FLEX(-6) (original expression AUX_ENDO_LAG_17_2(-1)) 
P(-2) (original expression P(-1)) 
P(-4) (original expression AUX_ENDO_LAG_15_1(-1)) 
P(-6) (original expression AUX_ENDO_LAG_15_2(-1)) 
P_REG(-2) (original expression P_REG(-1)) 
P_REG(-4) (original expression AUX_ENDO_LAG_16_1(-1)) 
P_REG(-6) (original expression AUX_ENDO_LAG_16_2(-1)) 
Y(-2) (original expression Y(-1)) 
Y(-4) (original expression AUX_ENDO_LAG_53_1(-1)) 
Y(-6) (original expression AUX_ENDO_LAG_53_2(-1)) 
P_US(-2) (original expression P_US(-1)) 
P_US(-4) (original expression AUX_ENDO_LAG_109_1(-1)) 
P_US(-6) (original expression AUX_ENDO_LAG_109_2(-1))

```

I also attach the model.  
Thank you!

[endo\_nonlinear\_loss\_LR\_est\_final1.mod](https://forum.dynare.org/uploads/short-url/xd6Io4zbyAeXcMAamlQdWONW8eV.mod) (63.2 KB)