The 1843 Closed-Form Trick That Solved Fibonacci
For 643 years after Leonardo Fibonacci described his rabbit-population sequence in 1202, computing the 100th Fibonacci number meant computing all 99 before it. Then in 1843, French mathematician Jacques Binet published a single formula that returned $F_n$ directly from $n$, using only the golden ratio and a power:
$$F_n = \frac{1}{\sqrt{5}}\left[\left(\frac{1 + \sqrt{5}}{2}\right)^n - \left(\frac{1 - \sqrt{5}}{2}\right)^n\right]$$
The leap from "compute the previous 99 terms" to "plug $n$ in and get the answer" is the entire point of an explicit formula — also called a closed-form formula or the general term.
An explicit formula for a sequence ${a_n}$ is a formula that expresses $a_n$ as a function of $n$ alone: $a_n = f(n)$. Given any term position $n$, you can compute $a_n$ in one step without computing $a_1, a_2, \ldots, a_{n-1}$ first.
Explicit vs Recursive — Two Different Formula Types
Sequences can be described two ways:
Recursive formula. Each term defined in terms of one or more previous terms. Example: $a_n = a_{n-1} + 4$, with $a_1 = 3$. To compute $a_{100}$, you compute $a_2, a_3, \ldots$ first.
Explicit formula. Each term defined directly from its position. Example: $a_n = 4n - 1$. To compute $a_{100}$, plug in $n = 100$: $a_{100} = 4(100) - 1 = 399$.
Both formulas describe the same sequence $3, 7, 11, 15, 19, \ldots$. The recursive form mirrors the construction; the explicit form returns the answer in one step.
Property | Recursive | Explicit |
|---|---|---|
Needs previous terms? | Yes | No |
Easy to derive from pattern? | Often easier | Sometimes harder |
Fast for large $n$? | No (linear time) | Yes (constant time) |
Common in computer science? | Yes (recursion) | Yes (formulas) |
How To Find The Explicit Formula — By Sequence Type
The recipe depends on the sequence type. The four below cover almost every Grade 9–11 problem.
Arithmetic sequences
Same difference $d$ between consecutive terms. Formula: $a_n = a_1 + (n - 1)d$.
Derivation. Start at $a_1$. Each step adds $d$. After $n - 1$ steps, you've added $d$ a total of $n - 1$ times: $a_n = a_1 + (n - 1)d$. The $(n - 1)$ comes from counting steps, not terms — the first term is $a_1$ with zero steps, not $a_1 + d$.
Geometric sequences
Same ratio $r$ between consecutive terms. Formula: $a_n = a_1 \cdot r^{n - 1}$.
Derivation. Same logic — start at $a_1$, multiply by $r$ each step, count $(n - 1)$ steps.
Quadratic sequences
The second differences are constant. Formula: $a_n = an^2 + bn + c$ — three coefficients, solved by plugging in any three known terms.
Fibonacci and Binet's formula
Recursive: $F_n = F_{n-1} + F_{n-2}$ with $F_1 = F_2 = 1$. Explicit (Binet's): the closed form above using the golden ratio $\varphi = \tfrac{1 + \sqrt{5}}{2}$.
Quick — Standard — Stretch: Three Worked Examples
Quick — find the explicit formula for $3, 7, 11, 15, 19, \ldots$
Arithmetic with $a_1 = 3$ and common difference $d = 4$.
$a_n = 3 + (n - 1)(4) = 3 + 4n - 4 = 4n - 1$.
Final answer: $a_n = 4n - 1$. Check: $a_1 = 4(1) - 1 = 3$. ✓ $a_5 = 4(5) - 1 = 19$. ✓
Standard (Wrong-Path-First) — find the explicit formula for the geometric sequence $5, 15, 45, 135, \ldots$
Wrong path. First instinct — recognise geometric with ratio 3, then write $a_n = 5 \cdot 3^n$.
Check: $a_1 = 5 \cdot 3^1 = 15$. But the first term is 5, not 15. Off by one. The wrong path forgot that the first term sits at $n = 1$ — and at $n = 1$, the exponent in the formula must produce $r^0 = 1$, not $r^1 = r$.
Correct method. The formula is $a_n = a_1 \cdot r^{n - 1}$, with the exponent $(n - 1)$ to account for steps from the first term. Here $a_1 = 5$, $r = 3$:
$$a_n = 5 \cdot 3^{n - 1}$$
Check: $a_1 = 5 \cdot 3^0 = 5$. ✓ $a_2 = 5 \cdot 3^1 = 15$. ✓ $a_4 = 5 \cdot 3^3 = 135$. ✓
Final answer: $a_n = 5 \cdot 3^{n - 1}$.
This is the most common off-by-one slip in Grade 10 algebra cohorts at our McKinney TX center — roughly five of every ten first attempts on geometric sequences use $r^n$ instead of $r^{n-1}$. The check above (plug in $n = 1$, confirm you get $a_1$) catches it every time.
Stretch — find the explicit formula for the quadratic sequence $1, 4, 9, 16, 25, \ldots$
The terms are perfect squares — but let's pretend we don't notice and derive the formula from the differences.
First differences: $3, 5, 7, 9, \ldots$ — not constant. Second differences: $2, 2, 2, \ldots$ — constant. So the formula has the form $a_n = an^2 + bn + c$.
Plug in three knowns: $a_1 = 1$, $a_2 = 4$, $a_3 = 9$.
$a + b + c = 1$
$4a + 2b + c = 4$
$9a + 3b + c = 9$
Subtract row 1 from row 2: $3a + b = 3$. Subtract row 2 from row 3: $5a + b = 5$. Subtract: $2a = 2$, so $a = 1$. Then $b = 3 - 3(1) = 0$. Then $c = 1 - 1 - 0 = 0$.
Final answer: $a_n = n^2$. The terms are indeed the perfect squares — which we now have a derivation for, not just a recognition.
Why Explicit Formulas Matter — From Population Biology To Algorithm Runtime
Explicit formulas show up wherever rapid lookup matters.
Compound interest. Balance after $n$ years at rate $r$: $A_n = P(1 + r)^n$. A geometric sequence — explicit form lets you compute the balance after 30 years in one operation rather than 30.
Algorithm complexity. When computer scientists say an algorithm runs in $O(n^2)$ time, they mean the operation count is a quadratic sequence in the input size — and the explicit formula is what predicts the runtime on a million-element input.
Population biology. A population doubling every generation is geometric: $P_n = P_0 \cdot 2^n$. Fibonacci's original rabbit problem is the most famous example; modern epidemiological models for virus spread use the same closed-form thinking.
Spectroscopy and physics. The energy levels of a hydrogen atom follow $E_n = -\tfrac{13.6}{n^2}$ eV — an explicit formula in $n$ that predicts the wavelengths emitted by hydrogen. Bohr derived it in 1913.
The destination of an explicit formula is the ability to jump — to predict what happens at the 1000th step without grinding through 999 of them.
Where Students Lose Marks On Explicit Formulas
Mistake 1: Using $n$ where you should use $n - 1$
Where it slips in: Arithmetic and geometric sequence formulas.
Don't do this: Write $a_n = a_1 + nd$ or $a_n = a_1 \cdot r^n$.
The correct way: The exponent and the multiplier of $d$ count steps from the first term, not the position. At $n = 1$ (the first term), zero steps have been taken — so the formula must reduce to $a_1$ at $n = 1$. The fix is the $(n - 1)$. The rusher who pattern-matches to "multiply by $n$" loses a point on every problem.
Mistake 2: Confusing the recursive formula with the explicit formula
Where it slips in: Test questions that ask for "the formula" without specifying.
Don't do this: Write $a_n = a_{n-1} + 4$ when the question asks for the explicit form.
The correct way: The recursive form references previous terms; the explicit form is in $n$ alone. If the test asks for $a_{100}$, the recursive form forces you to compute $a_2, a_3, \ldots, a_{99}$ first; the explicit form gives you the answer in one step. The memorizer who learned only the recursive form gets stuck on large-$n$ questions.
Mistake 3: Forgetting to verify the formula on the given terms
Where it slips in: Any derivation — particularly the quadratic case in the Stretch example.
Don't do this: Derive the formula, write it down, move on.
The correct way: Plug in $n = 1, 2, 3$ and confirm the formula returns the given terms. If even one fails, the derivation is wrong. The second-guesser who always checks is right to do so — verification takes 30 seconds and catches the off-by-one and the sign-flip every time.
The real-world version of Mistake 1 — the off-by-one error — has a name in software engineering: the off-by-one error, also called a fencepost error. It is the second-most-common bug in production code (after the null-pointer error) and has caused incidents from financial-reporting errors to GPS-track miscalculations. The fix in code is the same as in algebra: count steps, not positions.
Explicit vs Recursive — Side-by-Side Comparison
The two ways of describing a sequence sit next to each other in the table below — same sequence, two formulas, two strengths.
The Same Sequence, Two Ways
Sequence | Recursive Formula | Explicit Formula | Compute $a_{100}$ |
|---|---|---|---|
Arithmetic $3, 7, 11, 15, \ldots$ | $a_n = a_{n-1} + 4$, $a_1 = 3$ | $a_n = 4n - 1$ | $4(100) - 1 = 399$ in 1 step |
Geometric $5, 15, 45, 135, \ldots$ | $a_n = 3 a_{n-1}$, $a_1 = 5$ | $a_n = 5 \cdot 3^{n-1}$ | $5 \cdot 3^{99}$ in 1 step |
Quadratic $1, 4, 9, 16, 25, \ldots$ | $a_n = a_{n-1} + (2n - 1)$, $a_1 = 1$ | $a_n = n^2$ | $100^2 = 10{,}000$ in 1 step |
Fibonacci $1, 1, 2, 3, 5, 8, \ldots$ | $F_n = F_{n-1} + F_{n-2}$, $F_1 = F_2 = 1$ | Binet's: $F_n = \tfrac{\varphi^n - \psi^n}{\sqrt{5}}$ | One formula plug-in |
Strengths and Weaknesses — Side by Side
Property | Recursive | Explicit |
|---|---|---|
Definition | Each term defined from previous terms. | Each term defined directly from its position $n$. |
Easiest to spot from a pattern? | Often easier — just identify the rule connecting consecutive terms. | Sometimes harder — requires algebraic derivation. |
Time to compute $a_n$? | $O(n)$ — must compute all previous terms first. | $O(1)$ — constant time. One operation. |
Memory needed? | Stores all earlier terms (or the previous one or two). | Just the formula. |
Best for very large $n$? | Slow — computing $a_{1{,}000{,}000}$ takes a million steps. | Fast — plug in once. |
Programming idiom | Iteration / loops / recursion. | Direct formula evaluation. |
Closed form always exists? | Yes — the recursion itself is a form. | No — some recursive sequences (e.g., Collatz) have no known closed form. |
How to Convert Recursive → Explicit (by Sequence Type)
Sequence Type | Identify | Build the Explicit Formula |
|---|---|---|
Arithmetic | First term $a_1$ and common difference $d$ | $a_n = a_1 + (n - 1) d$ |
Geometric | First term $a_1$ and common ratio $r$ | $a_n = a_1 \cdot r^{n - 1}$ |
Quadratic | Constant second differences | Set up $a_n = An^2 + Bn + C$; solve $A, B, C$ from three known terms |
Linear recurrence (e.g., Fibonacci) | The characteristic equation | Roots of the characteristic polynomial → closed form (e.g., Binet's for Fibonacci) |
When to Use Which
Need just the next term? Either works. Recursive is more natural.
Need the 100th term? Use the explicit formula — one step, no iteration.
Modelling a real-world process? Recursive often mirrors the construction (population grows by $r$ each generation, balance compounds annually). Explicit is the answer to "what is the value at time $t$".
Programming? Both — but for performance-sensitive code, the explicit formula wins every time on large $n$.
Off-by-One — Where Recursive ↔ Explicit Most Often Breaks
The exponent and the multiplier of $d$ count steps, not positions. At the first term ($n = 1$), zero steps have been taken — so the explicit formula must reduce to $a_1$ at $n = 1$. The fix is the $(n - 1)$ that appears in both arithmetic and geometric explicit formulas; using $n$ instead shifts every term by one position. The numerical check (plug in $n = 1$, confirm $a_1$ comes back) catches the error in under five seconds.
The Five-Bullet Summary
An explicit formula gives the $n$-th term $a_n$ of a sequence directly as a function of $n$, without needing previous terms.
For arithmetic sequences: $a_n = a_1 + (n - 1)d$. For geometric: $a_n = a_1 \cdot r^{n - 1}$. For quadratic: $a_n = an^2 + bn + c$ (three coefficients from three knowns).
The single most common mistake is the off-by-one error — using $n$ instead of $n - 1$ in the exponent or multiplier.
Binet's formula (1843) is the most famous closed form — Fibonacci's recursive sequence expressed entirely in terms of $n$ and the golden ratio.
Explicit formulas matter wherever fast lookup matters — compound interest, algorithm runtime, energy levels of atoms.
Quick self-check — try these
If you get stuck on the off-by-one issue, return to the Standard worked example.
Find the explicit formula for the arithmetic sequence $7, 12, 17, 22, 27, \ldots$. Use it to find $a_{50}$.
Find the explicit formula for the geometric sequence $2, -6, 18, -54, \ldots$. Use it to find $a_{8}$.
Find the explicit formula for the quadratic sequence $2, 6, 12, 20, 30, \ldots$ by setting up three equations in $a, b, c$.
Want a live Bhanzu trainer to walk through more formula derivations? Book a free demo class — online globally.
Was this article helpful?
Your feedback helps us write better content