OUTLIERS and DUMMY VARIABLEs

HI all, I have a vector of returns of dimension 100x1 (obtained from the price of a financial time series) The boxplot of the series highlights some outliers. In particular, some outliers are linked to stock collapses and euphoria moments. I wanna fit a proper model for this time series.

How should I treat these outliers? I thought of using dummy variables. is that correct? How should I do?

Thanks in advance for your consideration and any help! Maty

[Matt Kindig]

Treat these outliers for what purpose? Its not clear what "treating" the outliers differently even means.

[Maty]

Thanks for your interest. I have rewritten the question.

Answer 1

[Jos (10584)]

See my answer on CSSM:

http://www.mathworks.nl/matlabcentral/newsreader/view_thread/330636