
Real Analysis Real Analysis Math Forum 
 LinkBack  Thread Tools  Display Modes 
January 26th, 2016, 12:50 PM  #1 
Member Joined: Jun 2009 Posts: 83 Thanks: 1  Approximation near the singularity
Hi, I would like to somehow "approximate" function f(x) near its singularity p ($\displaystyle f(p)=\infty$) by a sum of powers of x. To be more precise, let f be function from R to R (real numbers  but if it would be more convenient, let's use complex numbers instead), let f had as "neat" properties as you wish (eg. let f have onesided derivatives of any order) in the neighbourhood its singularity p (where $\displaystyle f(p)=\infty$). An example of such a function is $\displaystyle f(x):=1/x$ and p:=0 and we would like to approximate f(x) "near" x=0 (ie. where x>0 and x is "very small"). We would like to approximate f somehow by (from now we will suppose wlog that p=0) "sum" (*) "$\displaystyle \sum_{i=0}^{\infty}(a_i.x^i)$"  but in fact this is not (convergent) sum for any x. The question is how to precisely define, that the sum (*) approximates f (near p) well and how to find the coefficients a_i. Maybe that some methods of asymptotic analysis (eg asymptotic series) can be used but I would also like to see explicit construction (or general method) of the coefficients $\displaystyle a_i$ (say for functions 1/x and ln(x)  near 0). Thank you for any help. 
January 26th, 2016, 01:00 PM  #2 
Math Team Joined: Dec 2013 From: Colombia Posts: 7,600 Thanks: 2588 Math Focus: Mainly analysis and algebra 
Since the $i=0$ term is the dominant one for small $x$, your $a_i$ clearly do not have well defined values. What we tend to do with such functions is to determine the Laurent series, which includes negative powers of $(xp)$. Obviously, this doesn't do much for $f(x)=/frac1x$, but it's useful for Complex Analysis. 

Tags 
approximation, asymptotics, singularity 
Thread Tools  
Display Modes  

Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Residue at the essential singularity  aaronmath  Complex Analysis  3  October 16th, 2013 05:51 PM 
Singularity of a curve  mathbalarka  Calculus  5  April 14th, 2012 02:48 PM 
Pi Approximation  Wissam  Number Theory  16  March 13th, 2011 05:41 PM 
Linear Algebra  Singularity problem  bullroarer  Algebra  4  May 21st, 2009 12:16 PM 
f has nonisolated singularity. show that f is dense.  goldgrill  Complex Analysis  0  November 19th, 2007 12:34 AM 