SI
SI
discoversearch

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.
Technology Stocks : LSI Corporation -- Ignore unavailable to you. Want to Upgrade?


To: E. Graphs who wrote (10330)3/2/1998 2:54:00 AM
From: shane forbes  Respond to of 25814
 
E

I found the Bloomberg Personal article. It is in the Sept. '97 issue.

What they do (and they is not ML as I said but Dorsey, Wright & Associates) is:

(a) For a given stock look at a point-and-figure chart and see if the stock is indicating a "buy signal" or not.

(b) For all the stocks in a given sector (or a large enough representative group I guess) repeat (a).

(c) Now determine the "bullish percent" by dividing the stocks in the sector that show "buy signals" by the total number of stocks in the sector that you have looked at.

(d) Now repeat the above for each and every sector in the market.

(e) Thus after (d) we'll have "bullish percents" for all the sectors in the market.

(f) Plot all these "bullish percents" on a graph. If the graph is skewed right (that is most sectors have high "bullish percents" or are overbought) the market is overbought. Conversely if skewed left the market is oversold and ready to advance. A normal range is not skewed.
They use 30% bullish percent as a low level and 70% as a high level.

----

Compared to your method you are right this is the inverse! Still the outcome in terms of overall market condition are probably the same.

----

About programming it I'm sure it's possible. The difficult thing as I see it is how to first convert the chart reading instincts to mathematical rules! Next would be a simple way to download all the useful data! I don't mind thinking about this a bit more. It seems like it would be fun.

----

It would be interesting to see whether the Dorsey graph is skewed left, center or right these days. Your data implies that it is definitely not skewed right and may in fact be a bit skewed left - i.e., gains continuing. It would be cool to see if their data says the same thing!

----

Shane.