Hostname: page-component-8448b6f56d-dnltx Total loading time: 0 Render date: 2024-04-18T17:14:37.368Z Has data issue: false hasContentIssue false
  • English
  • Français

De-Trending Time Series Data for Variability Surveys

Published online by Cambridge University Press:  01 May 2008

Dae-Won Kim ,
Pavlos Protopapas  and
Rahul Dave
Dae-Won Kim
Affiliation:
Department of Astronomy, Yonsei University, Seoul, Korea email: dakim@cfa.harvard.edu Harvard Smithsonian Center for Astrophysics, Cambridge, MA, USA Initiative in Innovative Computing at Harvard, Cambridge, MA, USA
Pavlos Protopapas
Affiliation:
Harvard Smithsonian Center for Astrophysics, Cambridge, MA, USA Initiative in Innovative Computing at Harvard, Cambridge, MA, USA
Rahul Dave
Affiliation:
Initiative in Innovative Computing at Harvard, Cambridge, MA, USA
Rights & Permissions [Opens in a new window]

Abstract

Core share and HTML view are not available for this content. However, as you have access to this content, a full PDF is available via the ‘Save PDF’ action button.

We present an algorithm for the removal of trends in time series data. The trends could be caused by various systematic and random noise sources such as cloud passages, change of airmass or CCD noise. In order to determine the trends, we select template stars based on a hierarchical clustering algorithm. The hierarchy tree is constructed using the similarity matrix of light curves of stars whose elements are the Pearson correlation values. A new bottom-up merging algorithm is developed to extract clusters of template stars that are highly correlated among themselves, and may thus be used to identify the trends. We then use the multiple linear regression method to de-trend all individual light curves based on these determined trends. Experimental results with simulated light curves which contain artificial trends and events are presented. We also applied our algorithm to TAOS (Taiwan-American Occultation Survey) wide field data observed with a 0.5m f/1.9 telescope equipped with 2k by 2k CCD. With our approach, we successfully removed trends and increased signal to noise in TAOS light curves.

Type
Contributed Papers
Information
Proceedings of the International Astronomical Union , Volume 4 , Symposium S253: Transiting Planets , May 2008 , pp. 370 - 373
Copyright
Copyright © International Astronomical Union 2009

References

Everett, M. E. & Howell, S. B., PASP, 2001Google Scholar
Hammouda, K. M. & Kamel, M. S., IEEE Transactions on Knowdeledge and Data Engineering, Vol. 16, No. 10, 2004Google Scholar
Daniels, K. & Giraud-Carrier, C., Proceesing of 5th International Conference on Machine Learning and Applications, 2006Google Scholar
Grabmeier, J. & Rudolph, A., Data Mining and Knowledge Discovery, Vol. 6, No. 4, 2004Google Scholar
Jain, A. K., Murty, M. N., & Flynn, P. J., ACM Computing Surveys, Vol. 31, No 3., 1999CrossRefGoogle Scholar
Kim, et al. in preparation.Google Scholar
Lehner, M. J., Wen, C.-Y., Wang, J.-H., Marshall, S. L., Schwamb, M. E., Zhang, Z.-W., Bianco, F. B., Giammarco, J., Porrata, R., Alcock, C., Axelrod, T., Byun, Y.-I., Chen, W. P., Cook, K. H., Dave, R., King, S.-K., Lee, T., Lin, H.-C., Wang, S.-Y., arXiv:0802.0303Google Scholar
Mandel, K. & Agol, E., ApJL, 2002Google Scholar
Shi, J. & Malik, J., IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 22, No. 8, 2000Google Scholar
Young, A. T., Genet, R. M., Boyd, L. J., Borucki, W. J., Lockwood, G. W., Henry, G. W., Hall, D. S., Smith, D. P., Baliumas, S. L., PASP, 1991Google Scholar
Zhang, Z.-W. et al. in preparation.Google Scholar