Concept 05 / 10Sets, Relations and Functions
Functions: Definition, Domain and Range Techniques
A function as a relation giving every input exactly one output, the vertical line test, when two functions are equal, the rules for finding domains (denominators, even roots, logarithms, inverse trigonometric functions) and the standard methods for finding ranges.
A function is a relation from to in which every element of is related to exactly one element of . Two conditions are packed into that sentence: every element of must have an image (no element is left out), and no element may have two images. The set is the domain, is the codomain, and the set of images actually attained, , is the range. It is allowed for two inputs to share an output, and it is allowed for elements of to be unused.
For a real function drawn as a graph, 'exactly one output per input' becomes the vertical line test: a curve is the graph of a function of if and only if every vertical line meets it at most once. The parabola passes the test, while the circle fails it, because the vertical line meets it at . The circle can be split into two functions, and , which is precisely what happens when you solve an equation for and have to choose a sign.
Two functions are equal only if they have the same domain and give the same value at every point of that domain. The formula alone is not the function. So and are not equal, because is in the domain of but not of . Likewise and differ: the first is defined for all and equals , the second only for .
Finding the domain. When a function is given only by a formula, its domain is taken to be the largest set of real for which the formula produces a real number. Collect the conditions from every part of the expression and intersect them.
Denominator: must not be .
Even root : requires (and if the root is also in a denominator).
Logarithm : requires , and .
and : require .
and : require ; and : .
with a variable exponent: usually requires .
A nested expression is handled from the outside in. For , the logarithm needs and the square root needs . Because the base is less than , the logarithm is decreasing, so . The domain is . Forgetting that a base between and reverses inequalities is the classic error here.
Finding the range. There is no single method; choose according to the form.
Method 1, solve for : write , solve for in terms of , and find the for which a valid exists in the domain. For , we get , valid for every , so the range is .
Method 2, the discriminant method for : cross-multiply to a quadratic in whose coefficients involve , and demand a real root, so the discriminant is . Check separately the value of at which the coefficient vanishes, and check that the found does not make the original denominator zero.
Method 3, monotonicity: a continuous function that is increasing on has range . Use limits at open ends.
Method 4, completing the square or a known bound: ; ; and AM-GM gives for .
Method 5, substitution: set a sub-expression, find the range of first, then the range of the outer function on that set of .
Key Takeaways:
- Every element of the domain has exactly one image; codomain elements may be used many times or not at all.
- Range codomain.
- Vertical line test: a graph represents a function of if and only if every vertical line meets it at most once.
- Equal functions need equal domains, not just matching formulas.
- Domain rules: denominator , even root argument , log argument with base and , for and .
- With a log base between and , taking antilogs reverses the inequality.
- For quadratic over quadratic, use the discriminant in , then check the excluded values.
- Pitfall: 'simplifying' before finding the domain, for example cancelling in and wrongly including .
Brain-Teaser 1: Find the range of .
Solution: The denominator is never , so the domain is . Set and cross-multiply: . If , this quadratic in has a real root if and only if . Factorising the difference of squares gives , so . The value is attained at . So the range is , with the maximum at and the minimum at . The logic: a value is in the range exactly when the equation has a solution, and the discriminant is simply the condition for that.
Brain-Teaser 2: Are and the same function?
Solution: No. For , both roots must be real, so and , giving domain . For , only the product must be non-negative, so , giving . The two agree on , but is also defined on while is not. The identity holds only when ; for negative and the left side is not even real. Many 'find the domain' traps are hidden applications of algebraic rules outside the range where they hold.
Brain-Teaser 3: Find the range of .
Solution: Use substitution. Let , which takes every value in . Then takes every value in , and since is decreasing for , takes every value in . So the range is . The inner factor only changes how fast runs through its values, not which values it attains, so it has no effect on the range. Recognising which features affect the set of values, and which only affect how quickly they are traced out, is the conceptual point.
Practice this concept
Free · no account required · instant solutions