Gary
GaryVasco
Posts: 3,352
|
Post by Gary on Jun 9, 2016 19:17:11 GMT
Vasco,
Regarding Ch 5, Ex. 30(iii), how do we know that the equation is precise? What is the rationale for discarding the n > 4 terms? Is it because the coefficients become smaller and the Schwarzian converges? For a general smooth curve K, it can't be because derivatives go to zero after the fourth term. How do we know that the n > 4 terms become sufficiently small that they make an insignificant contribution to the error?
Gary
|
|
|
Post by Admin on Jun 10, 2016 7:14:55 GMT
Vasco, Regarding Ch 5, Ex. 30(iii), how do we know that the equation is precise? What is the rationale for discarding the n > 4 terms? Is it because the coefficients become smaller and the Schwarzian converges? For a general smooth curve K, it can't be because derivatives go to zero after the fourth term. How do we know that the n > 4 terms become sufficiently small that they make an insignificant contribution to the error? Gary Gary The right hand side is a "precise expression" for the difference between the fourth terms in the two series and the "is approximately equal to" sign is there because the difference between the series for $\mathcal{S}_C(z)$ and $\mathcal{S}(z)$ is the "precise expression" plus terms of smaller magnitude in $(z-a)^4$ and higher powers. We are not discarding them but just saying they are much smaller than the third term. This is because the series is of the form $c_0+c_1(z-a)+c_2(z-a)^2+c_3(z-a)^3+...$ and because we are using the series "in the vicinity of $a$", then $h=|(z-a)|$ is "small" (<1) and so each term will be an order of magnitude smaller than the previous term: $c_0+c_1h+c_2h^2+c_3h^3+...$ This does assume that the values of the $c_j$ are all roughly the same order of magnitude. We are also assuming that $z$ lies within the circle of convergence at $a$. Vasco
|
|
Gary
GaryVasco
Posts: 3,352
|
Post by Gary on Jun 10, 2016 14:18:24 GMT
No further questions.
|
|