I Markov Processes: Estimating Transition Probabilities

  • I
  • Thread starter Thread starter WWGD
  • Start date Start date
  • Tags Tags
    Transition
AI Thread Summary
Recent discussions focus on estimating transition probabilities for Markov matrices, emphasizing the need for sample data on single-step transitions. Participants highlight the application of Maximum Likelihood Estimation (MLE) for sampling, suggesting the use of sample proportions to determine transition frequencies between states. The calculation involves dividing the number of transitions from state i to j by the total transitions or states for discrete cases, while also considering continuous cases. Additionally, there is interest in mixture models that incorporate Black-Scholes methodologies. The conversation underscores the complexity and various approaches to accurately estimate these probabilities.
WWGD
Science Advisor
Homework Helper
Messages
7,724
Reaction score
12,877
Hi,
What are recent results on estimating transition probabilities for Markov matrices. I believe @BvU works in that general area?
 
Physics news on Phys.org
Sorry, but: no !
 
WWGD said:
Hi,
What are recent results on estimating transition probabilities for Markov matrices.
Based on what? Do you have sample data of single-step transitions?
 
Are there several options? Im not aware on what area /aspect to sample.I think , yes, iirc, they applied MLE to sample data, but I couldn't come up with the actual statistic .
 
Last edited:
FactChecker said:
Based on what? Do you have sample data of single-step transitions?
Edit: Sample proportions; number of times there has been a transition between states i and j, divided by total number of transitions/states ( for the discrete case)s, or anything else that's available for the continuous case .
 
Last edited:
WWGD said:
Edit: Sample proportions; number of times there has been a transition between states i and j, divided by total number of transitions/states ( for the discrete case)s, or anything else that's available for the continuous case .
I would divide by the number of samples that were in state i before the single step. Also, j=i should be included.
 
Do you know of mixture models that use Black Sholes?
 

Similar threads

I Desirable estimator properties...
Replies
7
Views
2K
MHB Optimizing Markov Chain Production System Throughput w/ Exponential Rate of 50
Replies
1
Views
2K
I Intransitive implication
Replies
6
Views
1K
I How can I compute Markov transition probabilities from given data?
Replies
7
Views
2K
A Question about a property of a matrix of transition probabilities
Replies
6
Views
2K
A Martingale, calculation of probability - Markov chain needed?
Replies
3
Views
2K
The infinite limits of the probability transition matrix for Markov chain
Replies
24
Views
4K
A Deriving Probability Amplitude from Markov Density Function
Replies
4
Views
2K
I How many tosses to flip a coin HHTH?
Replies
41
Views
7K
Dynamic transition probability matrix in markov chains
Replies
1
Views
5K

Hot Threads

Recent Insights

Back
Top