scholarly journals On convergence in distribution of the Markov chain generated by the filter kernel induced by a fully dominated Hidden Markov Model

2016 ◽  
pp. 1-67
Author(s):  
Thomas Kaijser
Keyword(s):  
Markov Chain ◽  
Markov Model ◽  
Hidden Markov Model ◽  
Hidden Markov ◽  
Convergence In Distribution ◽  
Filter Kernel

Hidden Markov Processes: Basic Properties

2014 ◽  
Author(s):  
M. Vidyasagar
Keyword(s):  
Markov Chain ◽  
Markov Model ◽  
Hidden Markov Model ◽  
Markov Processes ◽  
Viterbi Algorithm ◽  
Hidden Markov ◽  
Basic Properties ◽  
Deterministic Function ◽  
Hidden Markov Processes ◽  
Chain Type

This chapter considers the basic properties of hidden Markov processes (HMPs) or hidden Markov models (HMMs), a special type of stochastic process. It begins with a discussion of three distinct types of HMMs and shows that they are all equivalent from the standpoint of their expressive power or modeling ability: Type 1 hidden Markov model, or a HMM of the deterministic function of a Markov chain type; hidden Markov model of Type 2, or a HMM of the random function of a Markov chain type; and hidden Markov model of Type 3, or a HMM of the joint Markov process type. The chapter also examines various issues related to the computation of likelihoods in a HMM before concluding with an overview of the Viterbi algorithm and the Baum–Welch algorithm.

scholarly journals A Dependent Hidden Markov Model of Credit Quality

2012 ◽  
Vol 2012 ◽  
pp. 1-13 ◽  
Author(s):  
Małgorzata Wiktoria Korolkiewicz
Keyword(s):  
Markov Chain ◽  
Markov Model ◽  
Hidden Markov Model ◽  
Credit Rating ◽  
Hidden Markov ◽  
Credit Ratings ◽  
The State ◽  
Credit Quality ◽  
The Em Algorithm ◽  
Rating Process

We propose a dependent hidden Markov model of credit quality. We suppose that the "true" credit quality is not observed directly but only through noisy observations given by posted credit ratings. The model is formulated in discrete time with a Markov chain observed in martingale noise, where "noise" terms of the state and observation processes are possibly dependent. The model provides estimates for the state of the Markov chain governing the evolution of the credit rating process and the parameters of the model, where the latter are estimated using the EM algorithm. The dependent dynamics allow for the so-called "rating momentum" discussed in the credit literature and also provide a convenient test of independence between the state and observation dynamics.

scholarly journals The dynamics of social deprivation in Mexico

2021 ◽  
pp. 1-20
Author(s):  
José Carlos Ramírez
Keyword(s):  
Markov Chain ◽  
Markov Model ◽  
Hidden Markov Model ◽  
Similar Pattern ◽  
Social Deprivation ◽  
Hidden Markov ◽  
Sample Period ◽  
First Order ◽  
Order Markov Chain ◽  
Markovian Approach

This paper aims to model the dynamics of social deprivation in Mexico using a Markovian approach. First, we establish a scenario where a list of items characterizing social deprivation evolves as a first-order Markov chain under the sample period (2002-2012). Then, we estimate latent states and ergodic vectors of a hidden-Markov model to verify the strength of the conclusions drawn from such a scenario. After collecting results from both kinds of analyses, we find a similar pattern of impoverishment. The paper's conclusions state that the evolution of Mexico's deprivation profile may slightly worsen soon.

scholarly journals Probability Model Based on Cluster Analysis to Classify Sequences of Observations for Small Training Sets

2020 ◽  
Vol 8 (1) ◽  
pp. 296-303
Author(s):  
Sergey S Yulin ◽  
Irina N Palamar
Keyword(s):  
Markov Chain ◽  
Markov Model ◽  
Hidden Markov Model ◽  
Short Term Memory ◽  
Hidden Markov ◽  
Probability Model ◽  
Training Data ◽  
Training Dataset ◽  
Experimental Comparison ◽  
Self Organizing Map

The problem of recognizing patterns, when there are few training data available, is particularly relevant and arises in cases when collection of training data is expensive or essentially impossible. The work proposes a new probability model MC&CL (Markov Chain and Clusters) based on a combination of markov chain and algorithm of clustering (self-organizing map of Kohonen, k-means method), to solve a problem of classifying sequences of observations, when the amount of training dataset is low. An original experimental comparison is made between the developed model (MC&CL) and a number of the other popular models to classify sequences: HMM (Hidden Markov Model), HCRF (Hidden Conditional Random Fields),LSTM (Long Short-Term Memory), kNN+DTW (k-Nearest Neighbors algorithm + Dynamic Time Warping algorithm). A comparison is made using synthetic random sequences, generated from the hidden markov model, with noise added to training specimens. The best accuracy of classifying the suggested model is shown, as compared to those under review, when the amount of training data is low.

Prediksi Kurs Rupiah Terhadap Dolar Dengan FTS-Markov Chain Dan Hidden Markov Model

2019 ◽  
Vol 6 (1) ◽  
pp. 32-41
Author(s):  
Maria Titah Jatipaningrum ◽  
Kris Suryowati ◽  
Libertania Maria Melania Esti Un
Keyword(s):  
Fuzzy Logic ◽  
Markov Chain ◽  
Exchange Rate ◽  
Markov Model ◽  
Hidden Markov Model ◽  
Viterbi Algorithm ◽  
Hidden Markov ◽  
State Sequence ◽  
Small Rise ◽  
The Exchange Rate

Hidden Markov model is a development of the Markov chain where the state cannot be observed directly (hidden), but can only be observed, a set of other observations and combination of fuzzy logic and Markov chain to predict Rupiah exchange rate against the Dollar. The exchange rate of purchasing and exchange rate of saling is divided into four states, namely down large, down small, small rise, and large rise are symbolized respectively S1, S2, S3, and S4. Probability of sequences of observation for 3 days later is computed by forwarding and Backward Algorithm, determine the hidden state sequence using the viterbi algorithm and estimate the HMM parameters using the Baum Welch algorithm. The MAPE result exchange rate of purchase of FTS-Markov Chain is 1,355% and the exchange rate of sale of FTS-Markov Chain is 1,317%. The sequences of observation which optimized within exchange rate of  purchase is X* = {S3,S3,S3}, within exchange rate of sale is also X* = {S3,S3,S3}. Keywords: Exchange rate, FTS-Markov Chain, Hidden Markov Model

Sign in / Sign up

Export Citation Format

Share Document