0

I know this question might be a bit too broad, but I am looking for some pointers for self-study.

I am given a set of data for which I have to identify the trends, and potentially come up with some polynomial equation.

I don't have much experience with these tasks. After some superficial online research I found articles about curve fitting, but I don't know how it works and its basic principles. I see the concept dates back to earlier times, so I assume it is established and there are useful resources to learn it.

Can you point me to some "established" resources in the field so I can study on my own.

Unfortunately my attempts so far did not turn any useful lecture videos or go-to book in the field. I found a wiki page and an interesting site like this.

Kristof Tak
  • 229
  • 3
  • 8
  • As it stands it's probably too broad, as you guess already, you'll need to narrow it down into simpler questions such as a request for references on a particular topic. You probably *don't* want a polynomial equation, for a variety of reasons. If you have some functional form in mind, there's linear least squares and nonlinear least squares, there's GLMs and a host of other approaches. If you don't have a particular functional form in mind and just want smooth curves, there's various approaches to producing smooth relationships (including splines, local linear smoothing, and so on). – Glen_b Feb 01 '16 at 22:08
  • [This answer](http://stats.stackexchange.com/questions/100357/nonlinear-statistics/100420#100420) illustrates some smooth relationships, links to questions about them and so on. It might be a starting point for additional searches of our site which may help narrow your question a bit (via an edit). – Glen_b Feb 01 '16 at 22:12
  • @Glen_b thanks a lot for your response and linked answer. Can you point me to some introductory materials for curve fitting? – Kristof Tak Feb 02 '16 at 08:24
  • There are several references at that link. If the two Fox references (and the additional chapters) are too high a level you'll have to explain more about what you know. – Glen_b Feb 02 '16 at 09:38

0 Answers0