14.384
Time Series Analysis
Studies theory and application of time series methods in econometrics, including spectral analysis, estimation with stationary and non-stationary processes, VARs, factor models, unit roots, cointegration, and Bayesian methods. Enrollment limited.
Fall 2013
Topics
- Introduction to stationary time series _ARMA, limit theory for stationary time series, causal relationships, HAC_
- Frequency domain analysis _Spectra; filters; transforms; nonparametric estimation_
- Model selection and information _Consistent estimation of number of lags, discussion of non-uniformity and post-selection inferences_
- VAR _Definition, estimation: OLS, ML, Granger causality, impulse response functions and variance decompositions_
- Structural VARs _Identification, short term restrictions, long-term restrictions_
- VAR and DSGE models _World decomposition, fundamentality of shocks, do long-run restrictions identify anything_
- Factor model and FAVAR _Motivation, principal components, choosing number of static and dynamic factors, structural FAVAR, IV regression with factors_
- Asymptotic theory of empirical processes
- Univariate unit roots and near unit root problem _Unit root problem, unit root testing, confidence sets for persistence, tests for stationarity_
- Structural breaks and non-linearity _Testing for breaks with known and unknown dates, multiple breaks, estimating number of breaks_
- Multivariate unit roots and co-integration _Estimating cointegration relations, canonical form_
- Persistent regressors (prediction regression) _Limit theory, Stambaugh correction, nuisance parameter problem, conservative procedures, conditional procedures_
- GMM and Simulated GMM _GMM estimation and asymptotic theory, testing in GMM setting, simulated method of moments and time series specifics: estimation of covariance structure, initial condition problem, indirect inference_
- Weak IV _What is weak IV?, alternative asymptotic theory, how to detect weak IV, procedures robust to weak IV, unsolved problems._
- Kalman filter and its applications _State-Space models, time varying coefficients_
- ML estimation of DSGE _Stochastic singularities problem, misspecification and quasi-ML, identification_
- Identification and weak identification of DSGE
- Bayesian concepts
- Markov Chain Monte Carlo (MCMC) _Metropolis-Hastings, Gibbs sampler, data augmentation_
- Estimation of DSGE models using Bayesian methods
Readings
- [Hamilton] = Hamilton, James D. Time Series Analysis . Princeton University Press, 1994. ISBN: 9780691042893.
- [Hall and Heyde] = Hall, Peter, and C. C. Heyde. Martingale Limit Theory and Its Application . Probability and Mathematical Statistics. Academic Press, 1980. ISBN: 9780123193506.
- [Griliches and Intriligator] = Griliches, Zvi, and Michael Intriligator, eds. Handbook of Econometrics . Vol. 3. North Holland, 1986. ISBN: 9780444861870.
- [Lütkepohl] = Lütkepohl, Helmut. Introduction to Multiple Time Series Analysis . Springer-Verlag, 1993. ISBN: 9780387569406.
- SES #: 6–7 · TOPICS: VAR · READINGS: *[Hamilton] Chapters 10, and 11. *[Lütkepohl] Chapters 2, and 3 (2005). Watson, M. “Vector Autoregressions and Cointegration.” Chapter 47 in Handbook of Econometrics . Vol. 4. North Holland, 1999. ISBN: 9780444887665.
- SES #: 12 · TOPICS: Asymptotic theory of empirical processes · READINGS: *[Hamilton] Sections 17.1–17.3. [Hall and Heyde] Chapters 3, 4, and 5, and the Appendix.
- SES #: 19 · TOPICS: GMM and simulated GMM · READINGS: *[Hamilton] Chapter 14. *[DeJong and Dave] Chapter 7. [Canova] Chapter 5.
- SES #: 24 · TOPICS: Bayesian concepts · READINGS: *[Hamilton] Section 12.3.
Prerequisites & course information
Prereq: 14.382 or permission of instructor G (Fall) 5-0-7 units
Sources
- Course catalogue ↗2026–2027
- MIT OpenCourseWare ↗Fall 2013
- MIT OCW calendar ↗
- MIT OCW syllabus ↗
- MIT OCW readings ↗
- Historical MIT OpenCourseWare offering. Named reading entries and schedule topics are automatically extracted from the linked source pages.