# What's the purpose of the "predictions" argument in the pymc.sample\_posterior\_predictive function?
**URL:** https://discourse.pymc.io/t/whats-the-purpose-of-the-predictions-argument-in-the-pymc-sample-posterior-predictive-function/11619
**Category:** v5
**Tags:** modeling
**Created:** [March 15, 2023, 11:33pm UTC](https://discourse.pymc.io/t/whats-the-purpose-of-the-predictions-argument-in-the-pymc-sample-posterior-predictive-function/11619 "2023-03-15T23:33:34Z")
**Posts on this page:** 2
**Page:** 1
### Author:  [@jroberayalas](https://discourse.pymc.io/u/jroberayalas)
#### Post date: [March 15, 2023, 11:33pm UTC](https://discourse.pymc.io/t/whats-the-purpose-of-the-predictions-argument-in-the-pymc-sample-posterior-predictive-function/11619/1 "2023-03-15T23:33:34Z")
Hi,
Is there any reference that elaborates on the purpose of the `predictions` argument in the [sample\_posterior\_predictive](https://www.pymc.io/projects/docs/en/latest/api/generated/pymc.sample_posterior_predictive.html) function? By default it is set to `False` and I thought that I could use the values in `trace.posterior_predictive` to get my predictions. However, after diving into the [Forecasting with Structural AR Timeseries](https://www.pymc.io/projects/examples/en/latest/time_series/Forecasting_with_structural_timeseries.html) notebook, I found out the author uses `predictions=True` and `trace.predictions`, and this produces wider intervals in the prediction plots.
Any guidance is appreciated.
---
### Author:  [@jroberayalas](https://discourse.pymc.io/u/jroberayalas)
#### Post date: [March 22, 2023, 7:56pm UTC](https://discourse.pymc.io/t/whats-the-purpose-of-the-predictions-argument-in-the-pymc-sample-posterior-predictive-function/11619/2 "2023-03-22T19:56:58Z")
This question has been answered [here](https://community.intuitivebayes.com/t/what-s-the-purpose-of-the-predictions-argument-in-the-pymc-sample-posterior-predictive-function/1027/2?u=jroberayalas). For convenience, I’m sharing the same screenshot with the explanation:
