The Time magazine tries to use Markov Chain to model a series of events. For example, one of users' subscriptions may lo
Posted: Thu Apr 28, 2022 8:03 am
The Time magazine tries to use Markov Chain to model a series of events. For example, one of users' subscriptions may look like the following 1 2 3 4 6 7 8 9 10 Month Subscribed 5 0 From the above 10-month subscription history, we know that each user needs to make a decision of changing or not changing state. The following table lists the transformation of the subscription sequence into the transition states. To State From State Month pair 12 23 34 45 5 = 6 67 78 89 9 10 Observation Active/Disabled) A>A АЭА A →D D D D→D D A AD D A | AA 1. The probability of transition from one state to the other state (shown in the above table) can be described in the following figure. PASA Ро»o PAD A A D Ро»
Fill in the probabilities to the following transformation matrix. To State Active From State Disabled Active % % Disabled % % Note that the first cell PA->A means the probability we can observe the number of times with an Active state, the user remains Active, while PA->p is the probability the user switched to Disabled state. 2. For a user who is currently subscribing the magazine, what is the probability of staying active in 1 month and 2 months? Answer:
Fill in the probabilities to the following transformation matrix. To State Active From State Disabled Active % % Disabled % % Note that the first cell PA->A means the probability we can observe the number of times with an Active state, the user remains Active, while PA->p is the probability the user switched to Disabled state. 2. For a user who is currently subscribing the magazine, what is the probability of staying active in 1 month and 2 months? Answer: