normal distribution formula in c# ?

[afr0]

Hi All,
I'm trying to implement the NORMSDIST function of MS excel in the C#. I found its implementation from the following link.
http://office.microsoft.com/en-us/excel/HP052091941033.aspx

and I've written the following code for this i.e.

static double NORMSDIST(double z_score) 
        {            
            double z_ = - ( (z_score * z_score) / 2 ); 
            double normDist = (1 / ((Math.Sqrt(2 * Math.PI)))) * ( Math.Exp(z_) );

            return normDist;
        }

unfortunately this isn't giving me the correct result, can someone point out for me what m i missing in here?

thanks in advance.

[ddanbe]

Your code is correct but it gives the value if cumulative is false.
The excel example gives the value when cumulative is true.
I don't have your function in my Excel version, just simlated it with a similar function and setting mu to zero and sigma to one.
Hope this helps a bit.

[ddanbe]

Forgot to notice this was your first post!
Welcome to Daniweb afr0!
Try to use code tags when you post code.
I use underscores sometimes but, I would never use them in math formulas. It is confusing, I did not notice your unary minus at first.:|
You also use a few parenthesis to many.
This code (with code tags!) shows it(I think it is much more clearer to read):

static double NORMSDIST(double Zscore)
        {
            double Z = -(Zscore * Zscore) / 2;
            double normDist = (1 / Math.Sqrt(2 * Math.PI)) * (Math.Exp(Z));

            return normDist;
        }

[afr0]

I'm using 2007 office suite. but can you tell me how can i set cumulative to true. From the given formula that i provided the link i couldn't see any such flag. can provide that formula ?

thanks.

[ddanbe]

I have a 2003 version of Excel with this : NORMDIST(x,mean,stddev,cumulative) Cumulative can be true or false.
Your C# function returns the y value for x under the bell curve, so it is not cumulative.
If you want a cumulative function, you have to calculate the area under the bell curve from minus infinity up to the value of x. Perhaps you already have an algoritm to do that?

[ddanbe]

Perhaps you can do something(as a start) with this snippet
Just plug in your function and play with some parameters to get the results you want.