All about flooble | fun stuff | Get a free chatterbox | Free JavaScript | Avatars    
perplexus dot info
Discussion Forums
Login: Password:Remember me: Sign up! | Forgot password

Forums > Reference
This is a place to ask questions about math terminology, and to post links to other resources out on the web.
2003-11-25 01:41:23

I was using my TI-83 plus to graph y=x!. To my deep surprise, the calculator said that there is a .5! Wow! Besides that, it also seemed that there was a solution for (-.5)! and every other multiple of 1/2 above these. The calculator gave irrational numbers for all of them. I've been puzzling over this over the last few days. Can someone explain why you can take the factorial of halves, or, in the likeliness that it is much too complex, verify that it isn't just a bug in the calculator or my mind playing tricks on me?

2003-11-25 06:28:33
Re: .5!

Given that the definition of n! is 1x2x3x...xn, then 0.5! is undefined. What answer did your calculator give? If the calculator gave 0.5 as the answer, then it could be working on the basis of 0.5! = (1/2)! = 1!/2! = 1/2 (=0.5) What solutions does it give for other multiples of 1/2? eg does 2.5! = 60 (=5!/2!)

You could always have a look at the instructions to see if it gives any clue.

Popstar Dave
2003-11-25 07:01:23
Re: .5!

I've got a TI-83 as well, and for me I'm getting:
(1/2)! = 0.8862269255 (x)
(3/2)! = 1.329340388 (y)
(5/2)! = 3.32335097 (z)
As a note:
x/y = 2/3
y/z = 2/5
x/z = 4/15
So, yeah...

As for factorials of negatives, I get an error for all negative integer values and negative fractional values EXCEPT:
(-1/2)! = 1.772453851
Using the values calculated above, this value of (-1/2)! is double the value for (1/2)!.

I'm stumped.

2003-11-25 08:32:16
Re: .5!

The version of factorials extended to the real domain (and even the complex domain) exists in the Gamma function. For integers, n! = Γ(n+1). A graph of the Gamma function can be found at Mathworld. So the calculator in this instance is giving you Γ(n+1). As (n-1)! = n!/n, (-1/2)!, defined in this manner, is (1/2)!/(1/2), or 2 (1/2)!.

2003-11-25 08:50:09
Re: .5!

And of course the factorials for negative integers do not exist, as -1! would be 0!/0, or 1/0.

Brian Smith
2003-11-25 09:49:57
Re: .5!

The exact value of (1/2)! = Gamma(3/2) = (sqrt(pi))/2

The exact value of (-1/2)! = Gamma(1/2) = (Gamma(3/2))/(1/2) = sqrt(pi)

2003-11-25 18:15:59
Re: .5!

I thought it might be related somehow to advanced calculus! As long as it's not a bug, I'm happy. Thanks!

2003-11-25 18:38:28
Re: .5!

I can't figure out why they'd limit the fractional possibilities to multiples of 1/2 though.

2003-11-25 19:12:09
Re: .5!

You mean in the calculator? I was wondering that too. Maybe the real bug in calculator is that it can't calculate the factorial of smaller fractions. lol

Professor K at MCC
2004-03-04 19:19:15
Re: .5!

The reason that the TI-83 and probably others allows for increments of .5 is that it is required to calculate the Probability Density function for a t-distribution. In the formula one encounters both (n/2)! and [(n-1)/2]! Where n can be any positive integer. So of the two expressions in the factorials must be of the form #.5 and thus the TI wants to know that value and to allow people to calculate it if they have to. As was mentioned previously the Gamma function is what we use to extend the factorial and Gamma(n+1)=n!. Hope this helps.

Grade 11 student
2004-03-12 12:07:36
Re: .5!

I always thought it went like this:

4,5! = 4,5 * 3,5 * 2,5 * 1,5 * (0,5)!

The (0,5)! could be replaced with either (sqrt())/2 or just 0,886226925 if you wanted.

= pi so you don't confuse it with n.

Steve Royer
2004-06-17 21:20:35
Re: .5!

my 20 year old casio fx-570 gives an error for 0.5!
(i think its on its 4th set of batteries) my son had
to get a ti graphing job for school math, batteries last
about 3 months go figure

2004-06-26 01:13:28
Re: .5!

In Excel the factorial function is FACT(#), where # is the upper multiplier.
It appears to treat any decimal like the INT function before doing the factorial calculation; I note the same occurs if I use vulgar fractions eg (3 + 1/2).

2004-10-08 12:44:07
Re: .5!

So what about .6!?
Does that have a simply expressable exact value?
Does any rational number?

Brian Smith
2004-10-08 22:09:13
Re: .5!

First, .6! = Gamma(1.6)
One definition of the gamma function is: Gamma(z) = Integral {0,1} (ln(1/t))^(z-1) dt
So .6! = Integral {0,1} (ln(1/t))^.6 dt = .89351

I gave only five places since with my calculator, this formula calculated .5! to only 5 correct decimal places due to the behavior of ln(t) at the t=0 end of the integral.

Aspiring Novice
2005-10-31 15:37:11
Re: .5!

I am a sophmore in high school but my math teacher just mentioned the exact same problem, noting how it takes many math ideas to create a new one. The upper calculus 2 class is asked to solve (1/2)! he wouldn't tell us why but the answer is precisly sqrt(2)/2. So they do have a precise value. You are p[robably much more aware than i am of why this is but i think it should help.

Aspiring Novice
2005-10-31 15:38:40
Re: .5!

oops didn't read the above post, i guess some of ou were a bit ahead of me

Copyright © 2002 - 2018 by Animus Pactum Consulting. All rights reserved. Privacy Information