|
SPEAKER: |
|
|
TITLE: |
"Statistical Inference With Time Series Data" |
|
DAY: |
Wednesday, November 11, 1998 |
|
TIME: |
3:30 p.m. [Coffee in BSB-202 at 3:00 p.m.] |
|
PLACE: |
BSB-108 |
In no area of application other than time series, is there so much obsession for assuming normality and trying to prove that various statistics are asymptotically normally distributed. This has led to certain developments which are clearly flawed. The correct solutions can be sought in terms of Pearson curves. We illustrate this by considering the famous random walk problem, i.e., testing phi=1 in an AR(1) model. We then discuss the estimation of parameters, and hypothesis testing, in the context of a general autoregressive model. The MML estimators are derived and shown to be highly efficient and robust (due to umbrella ordering of the weights).
Dr. Moti L. Tiku received his Ph.D. (1964) and D.Sc. (1984) degrees from Aberdeen University, U.K. He is presently Professor Emeritus in the Department of Mathematics and Statistics at McMaster University and an Adjunct Professor at Wilfrid Laurier University. He has held positions as Visiting Professor in India, Singapore and South Africa. Dr. Tiku has published many papers and several books in the areas of mathematical statistics, multivariate analysis, robust inference and time series. His work has been extensively cited in numerous books and research publications.
Dr. Tiku's talk will be based primarily on the publications listed below. The reference Tiku and Wong (1998) has been placed on reserve at Thode Library (STATS 770: Statistics Seminar).
[1] Vinod, H.D. and Shenton, L.R. (1996), "Exact Moments for Autoregressive and Random Walk Models for a Zero or Stationary Initial Value," ECONOMETRIC THEORY 12, pp. 481-499.
[2] Tiku, M.L. and Wong, W.K. (1998), "Testing for a Unit Root in an AR(1) Model Using Three and Four Moment Approximations," COMMUNICATIONS IN STATISTICS - SIMULATION 27(1), pp. 185-198.
[3] Tiku, M.L., Wong, W.K. and Bian, G. (1999), "Estimating Parameters in Autoregressive Models in Non-Normal Situations: Symmetric Innovations," COMMUNICATIONS IN STATISTICS - THEORY AND METHODS 28(3), to appear.
[4] Vaughan, D.C. and Tiku, M.L. (1999), "Estimation and Hypothesis Testing for a Non-Normal Bivariate Distribution and Applications," JOURNAL OF MATHEMATICAL AND COMPUTER MODELLING: SPECIAL ISSUE, J. Gani and G. Haynatzki, Eds., to appear.
[5] Pearson, E.S. and Tiku, M.L. (1970), "Some Notes on the Relationship Between the Distributions of Central and Non-Central F," BIOMETRIKA 57, pp. 335-346.