Analysis Series: questions regarding the article
Author: 07nioe
Creation Date: 9/14/2014 6:15 AM
profile picture

07nioe

#1
Hi Eugene,

I have used the Visualizer Analysis Series to find enhancing filters for my strategy. The tool is a very good idea, because it makes it easier to quickly improve a System.

Anyway, I have two questions regarding the following page:
http://www2.wealth-lab.com/WL5WIKI/PVAnalysis.ashx

First: I have tried to rebuilt the example strategy. Unfortunately, I get always different (worse) results. I use 9,95% of equity with Nasdaq 100 stocks. What preferences you have choosen for the example?

Second: As a result a filter for CMO < -60 was applied. But in the Filter type pull down menu greather than was set. Is it correct? Could it not be possible to set less than corresponding with <-60?

Thank you very much.

Best
Niels
profile picture

Eugene

#2
Niels,

QUOTE:
First: I have tried to rebuilt the example strategy. Unfortunately, I get always different (worse) results. I use 9,95% of equity with Nasdaq 100 stocks. What preferences you have choosen for the example?

Do you consider the fact that the Nasdaq 100's composition itself has changed since then? After all, the article was rewritten for WL5 soon 6 years ago.

QUOTE:
Second: As a result a filter for CMO < -60 was applied. But in the Filter type pull down menu greather than was set. Is it correct? Could it not be possible to set less than corresponding with <-60?

Consider this example for illustrational purposes. Feel free to explore the possibilities discovered by the addin.
profile picture

Cone

#3
1) That article was written years ago (different set of Nasdaq stocks), probably used Yahoo! data, doesn't use [time-of-day] priority (required to create a duplicable limit system simulation when trades are dropped), and since it wasn't documented at the time, other settings can't be known. There's no sense in trying to duplicate those results. The purpose is only to provide an example of how to use Analysis Series.

2) I think the image presented is unfortunate because it doesn't match the discussion. Presumably, the results for "less than" -60 showed relatively a higher correlated return. In the case of the image showing results for "greater than" we can still see that the average results degrade somewhat (except above -20, which must have included an outlier).
profile picture

07nioe

#4
Okay, I got the Point! Thank you

Niels