Sums, rounding, powers, logs and trigonometry — the arithmetic that most other formulas are built on. This page lists all 80 of them; each links to a full lesson with an example you can copy and a practice file.
ABSABS returns a number without its sign: −250 becomes 250.ACOSACOS is the inverse of COS. It appears in the formula for distance between two places on Earth.ACOSHACOSH returns the inverse hyperbolic cosine (input 1 or more).ACOTACOT returns the angle (0 to π) whose cotangent is x.ACOTHACOTH returns the inverse hyperbolic cotangent (|x| > 1).AGGREGATEAGGREGATE is like SUBTOTAL with 19 functions and extra options — most usefully, ignoring error values such as…ARABICARABIC reads Roman numerals so you can sort “Part IX” correctly after “Part VIII”.ASINASIN returns the angle (in radians) whose sine is the given number.ASINHASINH returns the inverse hyperbolic sine.ATANATAN turns a slope (rise/run) into an angle. ATAN2 handles all four quadrants.ATAN2ATAN2 returns the angle of a point, handling all four quadrants. Note Excel’s order is (x, y) —…ATANHATANH returns the inverse hyperbolic tangent (input between -1 and 1).CEILINGCEILING rounds up to the nearest multiple — prices ending in 5 or 9, pack sizes. CEILING.MATH is…CEILING.MATHCEILING.MATH rounds up to the next multiple; FLOOR.MATH rounds down. Great for pricing and pack sizes.CEILING.PRECISECEILING.PRECISE rounds up to a multiple regardless of the signs of number and significance; negatives round toward zero.COMBINCOMBIN counts the ways to choose items when order doesn’t matter — teams from a class, lottery tickets,…COMBINACOMBINA counts selections where the same item can be chosen more than once — scoops of ice cream…COSCOS also expects radians. Seasonal models sometimes use SIN and COS of the month to capture yearly cycles.COSHCOSH returns the hyperbolic cosine (e^x + e^-x)/2: the shape of a hanging cable.COTCOT returns 1/TAN(x).COTHCOTH returns 1/TANH(x).CSCCSC returns 1/SIN(x).CSCHCSCH returns 1/SINH(x).DEGREESDEGREES converts radians back to degrees.EVENEVEN rounds away from zero to the next even integer — handy when items come in pairs.EXPEXP returns e^x and undoes LN. Use it to turn summed log returns back into an ordinary percentage.FACTFACT returns n! (1 × 2 × … × n). It grows absurdly fast: FACT(20) is already about…FACTDOUBLEFACTDOUBLE multiplies every other number: 7!! = 7 × 5 × 3 × 1.FLOORFLOOR rounds down to a multiple — full boxes you can ship, 15-minute billing blocks.FLOOR.MATHFLOOR.MATH rounds down to the nearest multiple — slabs, time buckets and whole units.FLOOR.PRECISEFLOOR.PRECISE rounds down to a multiple regardless of signs; negatives round away from zero.GCDGCD returns the largest number that divides all inputs — for simplifying ratios and finding pack sizes.INTINT rounds a number down to the nearest integer. On dates with times, it strips the time.ISO.CEILINGISO.CEILING always rounds towards positive infinity, whatever the sign of the multiple — useful when data mixes positive…LCMLCM finds the smallest number all inputs divide into — when two repeating schedules line up again.LNLN is the logarithm with base e. Finance uses it for continuously compounded (log) returns.LOGLOG answers “what power gives this number?” — how many doublings, how many digits.LOG10LOG10 tells you the order of magnitude. Charts of values ranging from hundreds to crores are easier to…MDETERMMDETERM returns the determinant. Zero means the equations have no unique solution.MINVERSEMINVERSE inverts a square matrix — with MMULT it solves simultaneous equations.MMULTMMULT multiplies two matrices. In practice it computes many weighted sums at once — quantities × price lists…MODMOD returns the remainder after dividing. Great for alternating rows, cycles and time differences over midnight.MROUNDMROUND rounds to the nearest multiple you choose — prices to ₹5, times to 15 minutes.MULTINOMIALMULTINOMIAL counts ways to divide items into groups of given sizes.MUNITMUNIT returns an identity matrix — 1s on the diagonal, 0s elsewhere. Handy for checking MINVERSE results.ODDODD rounds away from zero to the next odd integer — for example committee sizes that must avoid…PERCENTOFPERCENTOF returns a subset’s share of a total — mostly used inside GROUPBY and PIVOTBY to show %…PERMUTPERMUT counts ordered arrangements — podium positions, codes without repeated digits.PERMUTATIONAPERMUTATIONA counts ordered arrangements where items can repeat: a 4-digit PIN has PERMUTATIONA(10,4) = 10,000.PIPI returns 3.14159265358979, accurate to 15 digits. It takes no arguments but still needs the brackets.POWERPOWER raises a number to a power; the ^ operator does the same. Its most useful business job…PRODUCTPRODUCT multiplies all numbers in its arguments, ignoring text and blanks.QUOTIENTQUOTIENT returns only the integer part of a division; MOD returns the remainder.RADIANSRADIANS converts degrees, because SIN, COS and TAN in Excel expect radians.RANDRAND returns a new random number between 0 and 1 on every recalculation — useful for random samples…RANDBETWEENRANDBETWEEN returns a random integer between two numbers. It recalculates on every change — convert to values when…ROMANROMAN turns 1–3999 into Roman numerals — chapter numbers, annual-report years.ROUNDROUND rounds a number to a given number of digits. Unlike cell formatting, it changes the actual value…ROUNDDOWNROUNDDOWN removes digits without rounding up — for example to calculate completed years or whole units.ROUNDUPROUNDUP always rounds up (away from zero); ROUNDDOWN always rounds down. Useful for packing, billing and slabs.SECSEC returns 1/COS(x).SECHSECH returns 1/COSH(x).SERIESSUMSERIESSUM adds a power series a1·x^n + a2·x^(n+m) + ..., e.g. to approximate e^x or sin(x) with a…SIGNSIGN returns 1 for positive, −1 for negative and 0 for zero — neat for direction arrows and…SINSIN expects radians. Wrap degrees in RADIANS().SINHSINH returns the hyperbolic sine (e^x - e^-x)/2.SQRTSQRT returns the positive square root.SQRTPISQRTPI returns SQRT(x*PI()), which turns up in statistics and physics formulas.SUBTOTALSUBTOTAL calculates a sum, count, average and more while ignoring rows hidden by a filter. It also ignores…SUMSUM adds numbers, cell references and ranges. It ignores text and empty cells, which makes it the safest…SUMIFSUMIF adds values where one condition is true. For two or more conditions use SUMIFS.SUMIFSSUMIFS adds values where all conditions are true — the backbone of formula-based reports.SUMPRODUCTSUMPRODUCT multiplies ranges row by row and adds the products. Because it handles arrays natively, it doubles as…SUMSQSUMSQ squares each value and adds them up — the building block of error measures such as RMSE.SUMX2MY2SUMX2MY2 returns Σ(x² - y²) for two matching ranges.SUMX2PY2SUMX2PY2 returns Σ(x² + y²).SUMXMY2SUMXMY2 returns Σ(x - y)²: the squared error between forecasts and actuals.TANTAN links an angle to a slope. Radians again — wrap degrees in RADIANS().TANHTANH returns the hyperbolic tangent; also an activation function in neural networks.TRUNCTRUNC drops digits after a given point, moving towards zero. Unlike INT, TRUNC(−8.9) is −8.