Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Strategies & Market Trends : Ask Vendit Off-Topic Questions -- Ignore unavailable to you. Want to Upgrade?

To: J.K. who wrote (6317)	3/11/2005 10:45:26 PM
From: Walkingshadow	Read Replies (1) \| Respond to of 8752

More excellent questions.... Unfortunately, I think I have markedly less-than-excellent answers, it has been a very long time since my hard-core statistics days. The regression channels are not the same as linear regression, or multivariable logistic regression. I posted once about how they are calculated, I could get that to you if you like. The way I use regression channels is more like an art than a science. I look at the general chart, and ask where the long-term, medium-term, and short-term trends are. Then, I adjust the duration of the chart to be appropriate to the trend of interest. If we are talking about a short-term trend, generally this means the 3 month or 5 month daily chart (it makes no sense to examine the short-term trend in a 5 year chart, of course). If I am interested in longer term trends and where they might be going, I will lengthen the chart to usually at least one year, and for clarity sake, switch from daily to weekly. Similarly, I set the regression channel duration. It is most helpful to me to set the regression channel to short durations if I am interested in a short-term trend. But if I am interested in a medium or long term trend, then a short duration for the regression channels is inappropriate and could well be wildly misleading. Sometimes, I do both at the same time. Reid posted one of these for me last weekend. I wanted to see how the medium duration regression channel fit within the regression channel formed by the long-term trend. themarketwind.com Message 21108146 Message 21108242 So I guess you can see that the way I use the regression channels is not exact, and also open to question. So if you think there is something inappropriate or misleading about how these are used, that is certainly a fair topic for discussion, and I encourage you to post your thoughts. T