I read many threads on this group related to calculating the normal
distribution but I couldn't find what I needed exactly. Is there any
way in the HP 50g to calculate the value of standard normal
distribution(area under the curve) using(given) just the z-values? For
eg:- Using the tables, 0.4772 is the value of standard normal
distribution for a z-value of 2.0 regardless of what the mean,
standard deviation or the data values were although I know that z-
value is calculated using these parameters.
Thank you.
Regards,
Gurveer
Are you familiar with the UTPN command?
If you put 0, 1 and z on the stack, in that order, then the UPTN command
returns the upper tail probability.
This will give you 0.5 minus your value of 0.4772, when rounded to 4
decimal places).
The 0 and 1 are for the mean and standard deviation, which are 0 and 1,
respectively, for the z distribution.
You can create a program to get your value directly from the z-score.
For example:
\<< 0. 1. ROT UTPN NEG .5 + \>>
will take a positive value on the stack as the z-score and return the
probability of a score being between 0 and that z-score.
--
Virgil
Thanks a lot Virgil. That's exactly what I wanted. Yeah, I'm somewhat
familiar with the command UTPN and the concept you told makes sense.
That's all I need and that simple program just works awesome for me.
Thank you once again.
Regards,
Gurveer