Suppose that u(t) = 1/t^2+1. Find simpler functions f(t) and g(t) such that u(t) = f(g(t))
There are multiple answers to this sort of question. The most obvious one would be \[\begin{cases}f(t)=\dfrac{1}{t^2+1}\\ g(t)=t\end{cases}\] which works because \[u_1(t)=f(g(t))=f(t)=\frac{1}{t^2+1}\] but something tells me you're expected to use something different. Perhpahs \[\begin{cases}f(t)=\dfrac{1}{t}\\ g(t)=t^2+1\end{cases}\] so that \[u_2(t)=f(g(t))=f(t^2+1)=\frac{1}{t^2+1}\] Or you can use \[\begin{cases}f(t)=\dfrac{1}{t+1}\\ g(t)=t^2\end{cases}\] so \[u_3(t)=f(g(t))=f(t^2)=\frac{1}{t^2+1}\] The point is thata there are many ways to do this.
wow, there is more than one answer to this
Yes, you can even make the component \(f\) and \(g\) as complicated as you want, so long as the composition results in the same function.
so what I have to do is come up with f(t) and g(t) that is going to give f(g(t))
Yep
what if I had f(t) as 1 and g(t) as t^2+1?
No, that wouldn't work because \(f\) is a constant function. If \(f(t)=1\), it doesn't matter what you use for \(g\) because \[f(g(t))=f(t^2+1)=1\not=\frac{1}{t^2+1}\]
Think of it this way: if there's no \(t\) in the functional expression for \(f(t)\), then it won't work. The variable must be included.
so it would have to be f(t) = 1/t and g(t) = t^2+1
Yes
what is u(t) then?
that looks to me like a typo
\(u(t)\) will always be \(\dfrac{1}{t^2+1}\)
It's a more compact form of \(f(g(t))\)
the first answer you gave to me would be simpler... but I think we have to use the second answer
It's your choice. Like I said, many possible answers to this one.
And choosing the "simplest" one doesn't really make sense. How simple it is would depend on the context.
yes, you are right... my exam on this material is coming up tuesday and I don't feel ready
Well if my opinion means anything, I suppose I'd call my \(u_2(t)\) the simplest choice. Good luck.
ok I was looking over the u2t solution... f(t^2+1) is the same as 1(t^2+1)?
1/(t² + 1), yes
Since \(f(t)=\dfrac{1}{t}\), and \(g(t)=t^2+1\), then \(f(g(t))=f(t^2+1)=\dfrac{1}{t^2+1}\). Basically, you replace any \(t\) you see with \(t^2+1\).
ok I understand that part
You might benefit from being exposed to other examples. Say \(f(t)=t^2\) and also that \(g(t)=t^2\). Then, \[f(g(t))=f(t^2)=\left(t^2\right)^2=t^4\] Makes sense?
yes
so you find out f(t) and g(t) if they are not given and then you replace t with the f(t)
Be careful with how you phrase that. If you want to find \(f\) and \(g\) such that \(f(g(t))\) matches up with the desired function, then you replace \(t\) in \(f\) with \(g(t)\). However, if you want to match up \(g(f(t))\) with the given function, then you replace \(t\) in \(g\) with \(f(t)\).
so if I have f(t) = t^2+2t and g(s) = 2s+1 f(g(s)) = f(t^2+2t)=(2s+1)=t^2+2s+1?
I am probably off...
The variable of the resultant function should always be the one given in the "innermost" function; in this case, \(s\). So, \[f(t)=t^2+2t~~~~~~~~g(s)=2s+1\\ f(g(s))=f(2s+1)=(2s+1)^2+2(2s+1)=4s^2+8s+3\]
I am taking notes. You have a good way of explaining things.
It is often difficult to get concise notes out of a 45 minute class period.
Join our real-time social learning platform and learn together with your friends!