Let me just say upfront I'm not a mathematician, I'm rather looking for a practical answer to my question. I was wondering if there is a polynomial approximation for the function

$$\max(0,x)=\left\{\begin{array}{ll} x, x>0\\0, x \leq 0\end{array} \right.$$ I was thinking something like the sigmoid function could be useful and since it can be tweaked to output values in the range $[0,1)$, then the answer to my question would be the product $x * sigmoid(x)$, correct? If so, in order to increase accuracy, I would need a sigmoid function with steeper slope around 0, so something like $\frac{1}{1+e^{-kx}}$ for some $k\geq1$. The sigmoid function has then a well-known Taylor series approximation which I could compute in Mathematica.

  • 1
    $\begingroup$ Not polynomial approximations but try: Plot[Ramp[x], {x, -1, 1}, PlotStyle -> Thick] and Plot[x UnitStep[x], {x, -1, 1}, PlotStyle -> Thick]. $\endgroup$
    – Syed
    Feb 5 at 10:55
  • 1
    $\begingroup$ Plot[x LogisticSigmoid[x], {x, -10, 10}] $\endgroup$
    – Syed
    Feb 5 at 11:15
  • 1
    $\begingroup$ @Jimakos You suggestion looks reasonable, why are you not happy with it? $\endgroup$ Feb 5 at 14:10
  • 1
    $\begingroup$ I wonder how you are going to use this polynomial. (It's end-use would affect what form you want the approximation in, in my imagination.) The Weierstrass Approximation Theorem says there is an approximation within any tolerance you wish to specify, although the degree may be really large. Chebyshev approximation (like Flinty's answer) usually gives a good approximation to the best approximation possible of a given degree. There are efficient ways to compute and evaluate high-degree polynomial approximations without expressing the approximation explicitly in polynomial form. $\endgroup$
    – Michael E2
    Feb 5 at 16:11
  • 1
    $\begingroup$ Your function is already piecewise polynomial, why do you need a polynomial? If it is to get something smooth then you could use Sqrt[1/a+x^2]+x to approximate it, where larger a increase the fit but makes the derivatives more extreme $\endgroup$
    – Coolwater
    Feb 6 at 11:07

3 Answers 3


Limit[x LogisticSigmoid[k x], k -> Infinity] gives ConditionalExpression[x, x > 0] so give a large k and you'll get something close to your Max / Ramp function. Unfortunately a Taylor Series will have poor convergence.

You could use Chebyshev polynomials like this (where I'm using 10 terms, although you can increase this):

f[x_] := Ramp[x]
basis = Array[ChebyshevT[#, x] &, 10, 0];
expansion = 2/Pi Integrate[f[x] #/Sqrt[1 - x^2], {x, -1, 1}] & /@ basis;
approx = -expansion[[1]]/2 + expansion . basis;
Plot[{f[x], approx}, {x, -1, 1}]

chebyshev expansion

HornerForm[approx] gives a fairly compact expression: $$x \left(x \left(\left(x^2 \left(\frac{448}{45 \pi }-\frac{256 x^2}{63 \pi }\right)-\frac{80}{9 \pi }\right) x^2+\frac{40}{9 \pi }\right)+\frac{1}{2}\right)+\frac{1}{9 \pi }$$

Unfortunately, Chebyshev polynomials are only good approximations on the interval [-1, +1].

  • 2
    $\begingroup$ Rescale (or algebra I) can be used to map an interval $[a,b]$ back and forth between $[-1,1]$. (+1) $\endgroup$
    – Michael E2
    Feb 5 at 16:13

In addition to @flint's answer, one can "play dumb" and do something like this, finding the least squares difference fit over a range:

With[{f = Ramp, range = {-5, 5}, n = 10},
  With[{params = Array[c, n]},
   With[{eqn = FromDigits[params, x]},
    eqn /. Last@Minimize[
       Integrate[(f[x] - eqn)^2,
        Prepend[range, x]], params]]]] // Together

$$\frac{-21879 x^8+1261260 x^6-26276250 x^4+303187500 x^2+512000000 x+172265625}{1024000000}$$

Plot[{Ramp[x], %}, {x, -5, 5}]

enter image description here


Here's my way to play dumb: Chebyshev interpolation. It's almost equivalent to @flinty's result, but in a different form. (The two approach each other as the degree increases.)

{a, b} = {-1, 2};  (* Pick interval *)
deg = 10;          (* Pick degree *)
xx =               (* Chebyshev nodes *)
  Rescale[Sin[Pi/2 Range[-1. deg, deg, 2]/deg], {-1, 1}, {a, b}];
yy = Max[0, #] & /@ xx; (* y = f(x) *)

(* Barycentric is particularly appropriate and efficient here *)
approx = Statistics`Library`BarycentricInterpolation[
   xx, yy,
   "Weights" -> (* optional: increases accuracy slightly *)
     Table[(-1)^k, {k, 0, Length@xx - 1}], {1 -> 1/2, -1 -> 1/2}]];

Plot[{Max[0, x], approx[x]}, {x, -1, 2}]

enter image description here

You can evaluate the derivative (first derivative only) by passing a 1 as a second argument:

Plot[approx[x, 1], {x, -1, 2}]

enter image description here

  • $\begingroup$ My "play dumb" argument was that if you don't know basis functions or other useful facts you can always use brute force with Integrate and Minimize and hope for the best! ;) $\endgroup$
    – kirma
    Feb 5 at 16:39
  • 1
    $\begingroup$ @kirma I just meant use it [my code] as a black box. :) $\endgroup$
    – Michael E2
    Feb 5 at 16:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.