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.
Strategies & Market Trends : Befriend the Trend Trading
SPY 675.37-1.2%Nov 4 4:00 PM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Dr. Stoxx who wrote (17115)3/27/2001 9:18:18 AM
From: Ed Weider  Read Replies (2) of 39683
 
TC, briefly... I'll try to lay out the formula. Let me know if comes across clear.

First it measures the STDDev of ROC(5) on both series, the stock and the NAZ COMP
... then the following formula is worked, I use excel by the way.

It measures the price correlation of both series returning a factor between -1 and +1. +1 means perfect similar pattern, -1 means perfect opposite pattern. 0 would mean no direction at all when compared. Anyways I'm sure you're familiar with how that works.

So here's the formula, roughly explained. It's not the actually the code used in excel, just the concept.

(STDDev(ROC(symbol,5)/STDDEV(ROC(Index,5)))* Correlation(SymbolPrice, IndexPrice)

Yes this measures the stock's percentage return based on 5 periods compared to the index's return on 5 periods. And if it is in sync or not. That's where the correlation part comes in.

I then check for confidence in the correlation by using a function in excel called RSQ(). This returns how consistent the data is... if it was just a blip in the data that caused it to be volatile then the confidence would be low, but if it was consistent then the conf. would be high.

I prefiltered the list to only accept conf. greater than 80%. Effectively getting rid of most spikes.
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext