|
PureBytes Links
Trading Reference Links
|
Jim, Lionel, & All,
I am reminded of an old saying:
It's not the size or shape of the magicians stick but how he waves it that does the trick!
J.>From: "Lionel Issen"
>Reply-To: metastock@xxxxxxxxxxxxx
>To:
>Subject: Re: Lag In Moving Avg
>Date: Sat, 10 Feb 2001 16:06:57 -0600
>
>Since the moving average is a summation of a set of values, by its very nature it will lag. To reduce the lag we lose information. The moving averages in Metastock, and in all or almost all software, are statistically incorrect in that the moving average is shifted forward 1/2 period so that it ends at the right end of the chart. If you shift the moving average back 1/2 the number of periods in the moving average, you will eliminate the lag, except that you lose any moving average for the last 1/2 period, that is at the right end of the chart where you want it.
>
>MACD and other oscillators try to overcome this in various ways, with some success.
>
>Last year there were some formulas posted for zero-lag moving averages. I tried them out, and overall the results were poor. Many decades ago Curtiss Dahl suggested using a moving average and the same moving average shifted either forward or back several periods ( I forget which it was). Signals were given by a crossing of the 2 moving averages. I think that the MACD and its cousin the PO (price oscillator) are better.
>
>Try the different kinds of moving averages that are available in metastock. Some may give you the kind of information that you are looking for.
>Lionel Issen
>lissen@xxxxxxxxx
> ----- Original Message -----
> From: wooglin.org@xxxxxxxxxxx
> To: List MetaStock
> Sent: Saturday, February 10, 2001 10:54 AM
> Subject: Lag In Moving Avg
>
>
> Aside from Jurik's proprietary method, does anyone know of a way to formulate a way to get the lag out of a moving average.
>
> Thanks,
>
> Jim Barone
>
Get your FREE download of MSN Explorer at http://explorer.msn.com
|