Estimation of Hidden Markov Model

Abstract With the development of economy, estimation has gradually received attention. Economic performance is essential to a company, that's why data analyst is very popular. Since Dongfeng Motor Corporation is one of the magnate company in Chinese vehicle market, estimation the data of Dongfeng could be very meaningful. There are many methods used to estimate economic performance, in this thesis we mainly focus on Hidden Markov Model (HMM).First of all, the thesis introduces the basic concept of Markov Process and Hidden Markov model, including three classes of problems, evaluation, decoding, learning problems. Also, the thesis introduces the corresponding solution algorithms, which are Forward-Backward algorithm, Viterbi algorithm, Baum-Welch algorithm.Secondly, the thesis introduces a special case of HMM, named Poisson Hidden Markov Model (PHMM), including a very clear explanation of PHMM and parameter estimation.Thirdly, the thesis gives an example of economic performance estimation of Dongfeng Motor Corporation. Several data sets can be used to do the estimation and different models should be used with the different kinds of data. The example uses sales volume data to make estimation with continuous-time hidden Markov model.Finally, the thesis gives future work directions. The estimation of different data would be introduced in the last part. Potential applications of the Poisson Hidden Markov Models to estimate economic performance are proposed.