I have attached the simplified form of my data. The variable "Gender" will be the hidden state in our model and "Pregnancy Outcome" will be the observation. What is confusing me is that I don't have a sequence of events in my dataset. As you can see in my data "Performa# 4" i.e "Afifa Bibi" has 3 events of pregnancy i,e 1,2 and 3 where Gender (hidden state) is Son, Daughter and Daughter. Performa #5 is the next patient and her pregnancy events. So I have to apply model on this data where I have 5000 patients whose data I will observe. As you can see some patients has only 1 event of pregnancy and some has only two means every patient has different number of events.

1- Do I have to compute my HMM for every single patient?

2- How can I apply HMM when every patient has different number of events?

3- How can I calculate their Transition and Emission probabilities?