Hello,

I seem unable to find an example for this anywhere: how can a time-dependent variable be modelled within the framework of pymc3?

Currently, I want to model an inhomogeneous Poisson process. For the sake of simplicity, lets say

\lambda (t) = a *t + b

where a and b are parameters I wish to determine from some observational data.

My best attempt at implementing this so far is below. I am not sure how to include the time-dependence.

```
import numpy as np
import pymc3 as pm
import theano
import theano.tensor as tt
import matplotlib.pylab as plt
# Observation
a_actual = 1.3
b_actual = 2.0
t = np.arange(10)
obs = np.random.poisson(a_actual * t + b_actual)
plt.figure()
plt.plot(t, obs)
plt.show()
a1 = tt.scalar('a')
b1 = tt.vector('b')
y = tt.vector('y')
out = a1 * y + b1
func = theano.function([a1, b1, y], [out])
niter = 3000
# Model
with pm.Model() as model:
a = pm.Uniform(name='a', lower=0, upper=10)
b = pm.Uniform(name='b', lower=0, upper=10)
#Expected value
lam = pm.Deterministic('lam', func(a, b, t))
count = pm.Poisson(mu=lam, name='count', observed=obs)
#Run Monte Carlo
trace = pm.sample(niter)
pm.plots.plot_posterior(trace=trace["a"])
pm.plots.autocorrplot(trace=trace, varnames=["a"]);
pm.plots.plot_posterior(trace=trace["b"])
pm.plots.autocorrplot(trace=trace, varnames=["b"]);
plt.show()
```

Thank you for any help or links to examples.