Hey!

I’m working on trying to infer the standard deviation of a random walk from some observed data. However the data is irregularly spaced (there could be 0, 1, 2 observations at any given time point), so I’m not sure how to setup a bayesian random walk model for this. The data is generated by a process similar to the following,

```
T = 100
std = 1.0
mu_array = np.zeros(T)
t = []
obs = []
for i in range(1, T):
p = np.random.rand()
mu_array[i] = std*np.random.randn() + mu_array[i-1]
if p <= 0.1:
obs.append(std*np.random.rand() + mu_array[i])
obs.append(std*np.random.rand() + mu_array[i])
t.append(i)
t.append(i)
if p <= 0.3:
obs.append(std*np.random.rand() + mu_array[i])
t.append(i)
else:
continue
```

Any advice on how to model this would be greatly appreciated! Thanks!