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 : TA-Quotes Plus -- Ignore unavailable to you. Want to Upgrade?

To: Debra Orlow who wrote (9098)	3/25/1999 11:17:00 AM
From: TechTrader42	Read Replies (1) \| Respond to of 11149

I assume it's the SMA. Make sure the start day and end day are correct. I suppose you could try comparing results with avgvol function to those with: MovAvg(0,10,vol) and EMovAvg You could also write a scan that gets your own moving average, just to compare, by summing the volume for each day and dividing by the number of days. Brooke

To: Debra Orlow who wrote (9098)	3/25/1999 3:54:00 PM
From: Bob Jagow	Read Replies (1) \| Respond to of 11149

Plain old SMA, Debra. Did you forget that avgvol(0,-29) corresponds to movavg(0,30,vol)? Bob

To: Debra Orlow who wrote (9098)	3/25/1999 11:48:00 PM
From: John Schott	Read Replies (2) \| Respond to of 11149

AVGVOL errors in Aside from the comment already posted about (0,-29) really being 30 days, I've got another suggestion and a long comment. When you do the check test cited, you might also check for rounding errors in both QP2 and MSW. I mean if the volume on a day is actually 1240 or 1275, might the values by truncated or rounded (which way) in one of the two programs - and might not that be effected by settings in any options setups. The other comment is based on working with my local AAII Computerized Investing SIG (sort of a TA group in drag). You have to realize that stock market EOD price and volume data is ALWAYS WRONG, even when it is TOTALLY CORRECT. Sound's stupid I know, but consider: * Approximately 20% of the typicals days volume on the top two exchanges is OFF EXCHANGE, and little if any of this (especially NASDAQ@) is likely to be reported. This is trading in the private exchanges (Instinet, the biggest), trades crossed inside the brokers onw network or cleared internally (not so likely with NYSE stocks) and just errors and corrections in the brokerages. Some of the exchange corrections take months to be done and I would expect these never make it into the vendors updates. * Many stocks trade on the international exchanges worldwide and a lot of that must not be reported. I mean I can buy UIS and DCX stock on about a dozen worldwide exchanges - So how does a trade made in Asia when it is 2 AM in New York with a broker who isn't a member of NASDAQ or the NYSE ever make it - and where to. * You can execute a NYSE stock trade in places like the Philadelphia Exchange - so does this get reported. Now your and my mind reject the idea that we miss a stock in a scan by a few hunderd shares of error, no matter where it came from. But I think we should therefore consider that that also means some of the stocks we get hits on should not be there. Skipping over more examples, the end thought is that we should not take any scan output as scientific gospel - but only as an educated guess. It all comes back to an engineering problem between accuracy and precision. You can use our PCs to make calculations to microfine precision (QP2 plus good QP2 data), but if your data source does not exceed the accuracy (real world sources of error in the data), precision isn't going to help get better answers. More simply said this is an art, not quite a prercise science - so don't try too hard to make it that.