Dear Friends and Colleagues,
Multi-Dimensional Sphere Model (MDSM, 1995, 1996, 1997) is a new model based on multivariate space (m-space) originally designed for vegetation classification and succession trend analysis. A mutual fund investment was conducted from 1998-6-19 through 1999-1-29 to test this model. During the testing period, the average random investment (ARI) of all the 20 funds increased 2.45%, and the best fund in the 20 increased 18.98%; at the same time, the MDSM managed funds increased 25.82%! The conclusion is that the MDSM can be applied to multivariate time series analysis and forecast the trends for the multivariate systems. For data, trading record, and worksheet, please visit
http://www.mdsm.org/in_action/012999.html the Yearly Report(98) and linked files.Thank you for your attention,
T. Jay Bai, Ph. D.
MDSM Research
P.O.Box 272628
Ft. Collins, CO 80527
USA
Jbai@lamar.colostate.edu http://www.mdsm.orgReferences:
Bai, T. J., T. Cottrell, D.Y. Hao, T. Te, 1997: Multi-Dimensional Sphere Model and Vegetation Instantaneous Trend Analysis. Ecological Modelling, Vol.97 No.1-2, pp75-86
Bai, T. J., D. Hao, T. Cottrell, Tala, 1996: Land Condition Trend Analysis Using A Multi-Dimensional Sphere Model. Grassland of China, No.2, 1996, pp.35-46.
Bai, T. J., D. Hao, T. Te, 1995: Range Successional Trend Analysis of Fort Carson using MDSM. In Bo Li: Modern Ecology Seminars. Science Publishing House, Beijing.
Bai, T. J., 1999: An Application of Multi-Dimensional Sphere Model on Stock Market Ordination. Acta Scientiarum Naturalium Universitatis NeiMongol, Vol. 30, No.2 (in press).
The last three are in Chinese with English abstracts.