-1

Given the following parameters:

  • λ = (A,B,π).
  • A = the state transition matrix
  • A = { a[i][j] } = { P(state q[i] at t | state q[j] at t+1) },
  • B = the observation matrix and
  • π = the initial distribution.

It is correct the sentence below? (making explicit the relation between λ and A):

a[i][j] = P(state q[i] at t | state q[j] at t+1) =P(state q[i] at t | state q[j] at t+1, λ)

Please some help!

JMFS
  • 297
  • 1
  • 4
  • 11
  • An addition, even though it's old: Already the assumption is a bit odd. Normally you will have `a[i][j] = P(state j at time t | state i at time t-1)` – aufziehvogel Jun 18 '13 at 10:49

1 Answers1

0
P(state q[i] at t | state q[j] at t+1) =P(state q[i] at t | state q[j] at t+1, λ)

This expression would be

P(state q[i] at t | state q[j] at t+1) =P(state q[i] at t | state q[j] at t+1, λ) / P(λ)
smhx
  • 2,246
  • 18
  • 22
  • Thanks Nicko for your response. To better understand your answer, let me ask the following: The event (state q[j] at t+1) is independent of (λ)? and the event ((state q[i] at t) ∩ (state q[j] at t+1)) is independant of (λ) ? Thanks again. – JMFS Mar 02 '12 at 19:28