Andrew Gelman responds to this question on how to include prior information on quantities of interest, not just on parameters in your model with a bit of Stan Code.

Link: https://statmodeling.stat.columbia.edu/2019/08/23/yes-you-can-include-prior-information-on-quantities-of-interest-not-just-on-parameters-in-your-model/

```
model {
target += normal(y | a + b*x, sigma); \\ data model
target += normal(a | 0, 10); \\ weak prior on a
target += normal(b | 0, 10); \\ weak prior on a
target += normal(a + 5*b | 4.5, 0.2); \\ informative prior on a + 5*b
```

How would one code the last line in PyMC3? I know how to put as expression as the `mu`

parameter of a Normal dist, but how do you put these strong experiential priors on that parameter?