Estimating time series models by state space methods in Python: Statsmodels

• Abstract
• Introduction
• State space models
  • Kalman Filter
  • Initialization
  • State and disturbance smoothers
  • Simulation smoother
  • Practical considerations
  • Additional remarks
  • Example models
  • Parameter estimation
• Representation in Python
  • Object oriented programming
  • Representation
  • Additional remarks
  • Practical considerations
  • Example models
• Maximum Likelihood Estimation
  • Direct approach
  • Integration with Statsmodels
  • Example models
• Posterior Simulation
  • Markov chain Monte Carlo algorithms
  • Implementing Metropolis-Hastings: the local level model
  • Implementing Gibbs sampling: the ARMA(1,1) model
  • Implementing Gibbs sampling: real business cycle model
• Out-of-the-box models
  • SARIMAX
  • Unobserved components
  • VAR
  • Dynamic factors
• Conclusion
• References
• Appendix A: Installation
  • Dependencies
• Appendix B: Inherited attributes and methods
  • sm.tsa.statespace.MLEModel
  • sm.tsa.statespace.MLEResults
  • SimulationSmoothResults
• Appendix C: Real business cycle model code