# AutoCorrTransProbs and CrossSectionCovarCorr for FHorz

**URL:** https://discourse.vfitoolkit.com/t/autocorrtransprobs-and-crosssectioncovarcorr-for-fhorz/666
**Category:** Uncategorized
**Created:** [June 6, 2026, 2:18am UTC](https://discourse.vfitoolkit.com/t/autocorrtransprobs-and-crosssectioncovarcorr-for-fhorz/666 "2026-06-06T02:18:04Z")
**Posts on this page:** 4
**Page:** 1

<div class="post-metadata">

### Author: ![robertdkirkby](https://discourse.vfitoolkit.com/user_avatar/discourse.vfitoolkit.com/robertdkirkby/32/287_2.png) [@robertdkirkby](https://discourse.vfitoolkit.com/u/robertdkirkby)
#### Post date: [June 6, 2026, 2:18am UTC](https://discourse.vfitoolkit.com/t/autocorrtransprobs-and-crosssectioncovarcorr-for-fhorz/666/1 "2026-06-06T02:18:04Z")

</div>

I implemented three new commands for analysing FHorz models, to match the InfHorz versions that were recently added.

EvalFnOnAgentDist\_AutoCorrTransProbs\_FHorz.m

EvalFnOnAgentDist\_CrossSectionCovarCorr\_FHorz.m

and also  
EvalFnOnAgentDist\_AgeConditionalStats\_CrossSectionCovarCorr\_FHorz.m

these obviously have had some but not large amount of testing.

They cannot yet be used in the calibration/estimation commands, nor do they support simoptions.conditionalrestrictions, nor is there PType. (If anyone specifically wants one of the three, let me know so I can prioritize it.) Currently the autocorr and transprobs are just one period, but if you want, e.g., three-period autocorr let me know and I will implement; similarly simoptions.agegroupings is not yet supported.

---

<div class="post-metadata">

### Author: ![aledinola](https://discourse.vfitoolkit.com/letter_avatar_proxy/v4/letter/a/3ec8ea/32.png) [@aledinola](https://discourse.vfitoolkit.com/u/aledinola)
#### Post date: [June 8, 2026, 9:36am UTC](https://discourse.vfitoolkit.com/t/autocorrtransprobs-and-crosssectioncovarcorr-for-fhorz/666/2 "2026-06-08T09:36:02Z")

</div>

> [@robertdkirkby](#):
>
> They cannot yet be used in the calibration/estimation commands

Bu they can be included in custom stats, right? So the function for custom statistics could call `EvalFnOnAgentDist_CrossSectionCovarCorr_FHorz`, say, and its output could be included in the output from custom stats.

---

<div class="post-metadata">

### Author: ![robertdkirkby](https://discourse.vfitoolkit.com/user_avatar/discourse.vfitoolkit.com/robertdkirkby/32/287_2.png) [@robertdkirkby](https://discourse.vfitoolkit.com/u/robertdkirkby)
#### Post date: [June 8, 2026, 8:47pm UTC](https://discourse.vfitoolkit.com/t/autocorrtransprobs-and-crosssectioncovarcorr-for-fhorz/666/3 "2026-06-08T20:47:57Z")

</div>

> [@aledinola](#):
>
> Bu they can be included in custom stats, right?

Yes you can do this for now, but they should just be handled automatically and so that will happen at some point in near futuer.

---

<div class="post-metadata">

### Author: ![robertdkirkby](https://discourse.vfitoolkit.com/user_avatar/discourse.vfitoolkit.com/robertdkirkby/32/287_2.png) [@robertdkirkby](https://discourse.vfitoolkit.com/u/robertdkirkby)
#### Post date: [June 11, 2026, 4:03am UTC](https://discourse.vfitoolkit.com/t/autocorrtransprobs-and-crosssectioncovarcorr-for-fhorz/666/4 "2026-06-11T04:03:02Z")

</div>

EvalFnOnAgentDist\_AutoCorrTransProbs\_FHorz.m

now supports semiz and e shocks.
