Testing for Normality - Chi-squared goodness-of-fit

[coffeeconsumer]

I'm needing to test a data series, say the last 100 closing values, for its conformity to a normal distribution. My purpose for doing this is to scan for balance areas in a time-volume profile.

I am considering using a Chi-squared goodness of fit test, however the programming of this looks quite daunting to me.

I'm just wondering if there are any native functions for statistical testing of normality? I've searched the MSDN Library and I came up with nothing.

Alternatively, has anyone already programmed such a test who might care to post it?

Cheers,
C.C.

[NinjaTrader_Dierk]

>> I'm just wondering if there are any native functions for statistical testing of normality?
None I'm aware of.

[darthtrader]

That is a very interesting idea. From my limited research on C# and stats it seems you would have to buy a 3rd party library most likely unless you want to do alot of leg work.
this one looks pretty interesting
http://www.centerspace.net/products.php?page=3

I'm going to try to make a simple correlation strategy this weekend that just spits out numbers to the output window between two instruments. Certainly post anything you find in this area, this is totally the kind of stuff i'm interested in. Once i've learned enough in ninja though i'm going to buy one of these stat libraries. I'm sure it would be worth the time saved programming wise.

Out of curiousity I wonder if at some point in the future the ninja guys could do a simple example of including a 3rd party library. Just something simple and open source to get the ball rolling.

[patrf]

I ended up on this page when searching for a test of normality. Eventually I found one here in case someone else is looking for one:
http://in-silico.net/statistics/lillieforstest

[NinjaTrader_Matthew]

patrf,

Thank you for sharing and welcome to our forums!