A pizza shop charges $$10$ flat delivery plus $$2$ per kilometre. For one customer's order, those two numbers are fixed. For the next customer, the formula stays the same but the numbers could change.
A parameter is a quantity that characterises a family of equations — held fixed within any one equation, but changeable from one equation to the next.
The Formal Definition
In mathematics, a parameter is a quantity that influences the output or behaviour of a mathematical expression but is treated as constant in a particular context. Variables are quantities that change within the same context.
The line equation $y = mx + b$ has:
$x$ and $y$ as variables (they vary along the line).
$m$ and $b$ as parameters (they pick which line we're on; once picked, they don't change).
Change $m$ and $b$, and you get a different line. Hold them constant, and you're tracing one specific line.
The same idea generalises: in a circle $x^{2} + y^{2} = r^{2}$, the radius $r$ is a parameter (each value of $r$ picks one specific circle). In statistics, a population mean $\mu$ is a parameter — a single number describing the entire population.
Quick reference.
Definition: a quantity held constant inside one equation but adjustable across a family.
Symbol/Notation: typically letters like $m, b, k, c, \mu, \sigma$.
Distinguished from: variables (which change inside the equation).
Used in: function definitions, parametric equations, statistics, regression.
Grade introduced: CCSS-M HSF-IF.C.7 (function family analysis); NCERT Class 11 — Statistics.
Parameter vs Variable — The One Distinction That Matters Most
The two words look interchangeable; they aren't. The difference is what you're allowed to change in the current context.
Variable | Parameter | |
|---|---|---|
Role | Changes inside the equation | Fixed inside the equation |
Across the family | The same variables work for the whole family | Different parameters pick different family members |
Example: $y = mx + b$ | $x, y$ | $m, b$ |
Example: $f(x) = a x^{2}$ | $x$ | $a$ |
Example: $\sigma$-normal $N(\mu, \sigma^{2})$ | the random variable | $\mu, \sigma$ |
A parameter today might be a variable tomorrow. If you're fitting a line to data and solving for $m$ and $b$, those become the unknowns — and $x, y$ become known data. The roles depend on context, not on the letter.
The Three Most-Used Settings for Parameters
1. Parametric equations
A parametric equation writes both $x$ and $y$ as functions of a third variable — the parameter — usually called $t$.
To trace a circle of radius $r$:
$$x = r \cos t, \quad y = r \sin t, \quad t \in [0, 2\pi).$$
Here $t$ is the parameter you vary; the circle is the trace it produces. To trace a line through two points $(x_{0}, y_{0})$ and $(x_{1}, y_{1})$:
$$x = x_{0} + t(x_{1} - x_{0}), \quad y = y_{0} + t(y_{1} - y_{0}), \quad t \in [0, 1].$$
Parametric equations are the standard way to describe curves in motion — projectile paths, planetary orbits, the trajectory of a robot arm.
2. Parameters in statistics
A statistical parameter is a single number describing a population.
Population mean $\mu$ — the true average of every member of the population.
Population standard deviation $\sigma$ — the spread.
Population proportion $p$ — the true fraction with a given property.
In contrast, a statistic is what you compute from a sample of that population — the sample mean $\bar{x}$, the sample standard deviation $s$. Statistics estimate parameters. Parameters are usually unknown; statistics are usually all we have.
3. Parameters in function families
A function family is a collection of functions sharing the same algebraic shape but differing in parameters.
Linear: $f(x) = mx + b$ — two parameters $m, b$.
Quadratic: $f(x) = ax^{2} + bx + c$ — three parameters $a, b, c$.
Exponential: $f(x) = A e^{k t}$ — two parameters $A, k$.
Each setting of the parameters picks one specific member of the family. The graph of every linear function looks like a line; the parameters $m, b$ decide which line.
Three Worked Examples — Quick, Standard, Stretch
Quick. In the function $h(t) = 20t$ (a tree growing $20$ cm per year), identify the variable and the parameter.
$t$ is the variable (it changes — the height depends on the year). $20$ is the parameter — it characterises this particular tree's growth rate. A different tree might have $h(t) = 15t$.
Final answer: variable $= t$; parameter $= 20$.
Standard (Wrong Path First — The Mistake Worth Making Once). Identify all the parameters and all the variables in the line equation $y = 3x + 7$.
The wrong path. A student writes: "$y, x, 3, 7$ are all variables because they're all letters or numbers."
The flaw: roles are assigned by function, not by being a letter or a number. $y$ and $x$ vary together along the line; $3$ and $7$ are fixed numbers that picked out this specific line.
The rescue. In the line $y = 3x + 7$:
$x$ varies (input).
$y$ varies with $x$ (output).
$3$ is the slope — a parameter.
$7$ is the $y$-intercept — a parameter.
Final answer: variables $= x, y$; parameters $= 3, 7$.
The lesson — a parameter is fixed inside the equation but defines which equation you're using. Changing $3 \to 5$ gives a different line. Changing $x$ traces points on the same line.
Stretch. Convert the parametric equations $x = 3\cos t, , y = 3\sin t$ into a single equation in $x$ and $y$ (eliminate the parameter $t$).
Square both: $x^{2} = 9\cos^{2} t$ and $y^{2} = 9\sin^{2} t$.
Add: $x^{2} + y^{2} = 9(\cos^{2} t + \sin^{2} t) = 9 \times 1 = 9$.
Final answer: $x^{2} + y^{2} = 9$ — a circle of radius $3$ centered at the origin.
The parameter $t$ has been eliminated; the resulting equation describes the same curve but no longer carries the "direction of tracing" information that $t$ provided.
Where Parameters Appear — Beyond the Textbook
A few places this idea quietly does heavy lifting:
Computer programming. Function parameters in code are exactly the same idea — a function
def area(r): return pi * r**2takes a parameterrand produces an output. The mathematical use is older but identical in spirit.Linear regression. Fitting a line to data means solving for the parameters $m$ and $b$ that minimise the squared error. This is the foundation of least-squares regression, introduced by Adrien-Marie Legendre (1752–1833, France) in 1805 and Carl Friedrich Gauss shortly after.
Modelling. Every population model, weather model, or economic model has parameters that are fit to data — growth rate, interest rate, transmission rate.
Machine learning. A neural network with millions of weights has millions of parameters. Training the network means searching for the parameter values that fit the data best.
The word parameter itself comes from Greek para- ("alongside") + metron ("measure") — literally, "the thing measured alongside the variable."
Tripping Points to Avoid in Parameter
Mistake 1: Calling every letter a variable
Where it slips in: A student labels $m$ and $b$ in $y = mx + b$ as variables.
Don't do this: Conflate "letter in the equation" with "varies in the equation."
The correct way: Within one specific line, $m$ and $b$ are fixed numbers — parameters. They vary across the family of lines, not within one.
Mistake 2: Mixing up parameter (population) with statistic (sample)
Where it slips in: A student writes the sample mean and calls it $\mu$.
Don't do this: Use a parameter symbol for a sample quantity.
The correct way: $\mu$ is the population mean (parameter); $\bar{x}$ is the sample mean (statistic). The sample statistic estimates the population parameter — they are not the same number.
Mistake 3: Eliminating the parameter without checking the curve traced
Where it slips in: Converting parametric $x = \cos t, y = \sin t$ to $x^{2} + y^{2} = 1$ without realising the parameter $t$ also carried direction and speed information.
Don't do this: Treat the elimination as lossless.
The correct way: Elimination gives you the shape of the curve. The parametric form additionally tells you how the curve is traced — clockwise or anti-clockwise, fast or slow. For motion problems, keep the parametric form.
A real-world version of the mistake. In linear regression, a parameter mistake cost a high-profile 2008 financial model — the "Gaussian copula" used to price mortgage-backed securities — most of its reliability when one parameter (the default correlation $\rho$) was mistakenly held constant across radically different housing markets. When the actual correlation drifted, the model's predictions held until they failed catastrophically. Knowing which numbers in your model are parameters (settable) versus constants (truly fixed) is not a textbook nicety.
Conclusion
A parameter is a quantity held constant inside an expression while variables change.
In $y = mx + b$, $x$ and $y$ are variables; $m$ and $b$ are parameters.
Parameters appear in parametric equations, statistical descriptions of populations, and every function-family analysis.
Changing a parameter changes which member of a family you're working with; changing a variable moves you within the same equation.
Confusing parameters with variables (or with sample statistics) is the most common slip.
Quick Self-Check — Three Problems
In the parabola $y = ax^{2}$, name the variable and the parameter.
The line $y = -2x + 5$ passes through which $y$-intercept? Is $5$ a variable or a parameter?
Eliminate the parameter from $x = 2t, , y = t^{2}$ to write $y$ as a function of $x$.
If problem 2 gave "variable," return to Mistake 1 above.
Want a live Bhanzu trainer to walk your child through function families and parametric equations? Book a free demo class — online globally.
Was this article helpful?
Your feedback helps us write better content