aka "Bayesian Methods for Hackers": An introduction to Bayesian methods + probabilistic programming with a computation/understanding-first, mathematics-second point of view. All in pure Python ;)
(I'm not sure if it's an issue with the book or I'm just plain stupid. Hope it's OK to post this as an issue).
I'm reading the book, and I'm having trouble at Chapter 1, Inferring behaviour from text-message data.
Let's assume that on some day during the observation period (call it τ), the parameter λ suddenly jumps to a higher value.
Right after this, I cease to understand the meaning of the words "posterior" and "prior" in this context. From this chapter, I understand that "prior" means the meaning of the variable before observing a certain event, and "posterior" - after observing it and changing our expectations about further events of this nature.
In this case, it would seem that event in question is the day τ; but it would mean that λ1 is the prior and λ2 is the posterior. However, further it's written about prior and posterior values of thees λN, which makes me completely confused - what event do we observe and on what timeline that changes the values of λ1 and λ2?
(I'm not sure if it's an issue with the book or I'm just plain stupid. Hope it's OK to post this as an issue).
I'm reading the book, and I'm having trouble at Chapter 1, Inferring behaviour from text-message data.
Right after this, I cease to understand the meaning of the words "posterior" and "prior" in this context. From this chapter, I understand that "prior" means the meaning of the variable before observing a certain event, and "posterior" - after observing it and changing our expectations about further events of this nature.
In this case, it would seem that event in question is the day τ; but it would mean that λ1 is the prior and λ2 is the posterior. However, further it's written about prior and posterior values of thees λN, which makes me completely confused - what event do we observe and on what timeline that changes the values of λ1 and λ2?