>
>Two Questions
>
>I can't solve 1.1 #24 c, d, or e. I can compute the actual value of erf(1)
>with my computer, and compare to values of the approximation with increaseing
>accuracy unitl I reach 10^-7.  I cannot figure out how to use the error
>formula to make this work.   
>
    1.1, part c: What you need to observe is the SIGN of the error terms, and
the relative size of the terms. I don't think that that's giving away too
much....
    Then part d follows, and you're to draw your conclusions in e.

>
>When using the error formula, you need to use the value for (That squiggley
>line ->"sq") that maximizes the error. This may not occur at x-nought, or
>x. It may also be at a different location for each derivative. So you need to
>find the critical points of the (N+1)st derivative and find the value at each
>and the value at the end points to find the maximum value. This seems like a
>lot of work. Am I overlooking something? 
>

    While it may seem like a lot of work, it's the work that must be done to
get the tightest error bound! Sorry!;) You're basically saying "how badly can
that next higher derivative behave on the interval", and using its bound,
because all Taylor's theorem guarantees is that it won't be any WORSE than
that....