PsychLib is a TypeScript library for math, statistics, and data analysis. Featured in psychological and educational research.
- PsychLib can be used in all modern JavaScript/TypeScript environments, including browsers, Node.js, Deno, and Bun.
- For use cases, please refer to my another project PsychPen.
For full documentation, see https://jsr.io/@psych/lib/doc.
npx jsr add @psych/lib # if using npm
bunx jsr add @psych/lib # if using bun
deno add jsr:@psych/lib # if using deno
pnpm dlx jsr add @psych/lib # if using pnpm
yarn dlx jsr add @psych/lib # if using yarnimport { randomNormal, WelchTTest } from '@psych/lib'
const groupA = new Array(500).fill(0).map(() => randomNormal(10, 2))
const groupB = new Array(500).fill(0).map(() => randomNormal(11, 2))
const tTest = new WelchTTest(groupA, groupB)
console.log(tTest.t, tTest.df, tTest.p) // Maybe -6.5..., 997.4..., 0.0...For full documentation, see https://jsr.io/@psych/lib/doc.
If you haven't installed deno yet, please install it referring to the https://deno.com. Then, clone this repository.
git clone https://github.com/LeafYeeXYZ/PsychLib.gitNow you can write TypeScript code in /lib/**/*.ts and export functions in /lib/index.ts. Note that you should not import base functions from /lib/index.ts to avoid circular dependencies. Instead, you can import them directly from /lib/base.ts or /lib/xxx/index.ts.
After writing the code, remember to add test cases in /tests/*.test.ts. You can run the test cases using the following command.
deno test -AYou can also add benchmark cases in /benchs/*.bench.ts and run the benchmark using the following command.
deno bench -AThis project publishes to https://jsr.io, so you don't need to compile the code to JavaScript. And you also don't need to publish the package manually. Just modify deno.json and push the code to the repository. The GitHub Action will do the rest for you.
CPU | Apple M3
Runtime | Deno 2.2.3 (aarch64-apple-darwin)
benchmark time/iter (avg) iter/s (min … max) p75 p99 p995
--------------------------------------------------- ----------------------------- --------------------- --------------------------
sum() - n=500 273.9 ns 3,652,000 (270.6 ns … 286.9 ns) 275.3 ns 282.1 ns 286.9 ns
mean() - n=500 275.8 ns 3,626,000 (271.3 ns … 303.9 ns) 276.2 ns 292.6 ns 303.9 ns
max() - n=500 178.0 ns 5,619,000 (171.9 ns … 197.9 ns) 185.1 ns 192.3 ns 195.9 ns
min() - n=500 175.0 ns 5,715,000 (169.4 ns … 194.1 ns) 181.8 ns 186.5 ns 187.2 ns
median() - n=500 6.9 µs 144,000 ( 6.0 µs … 11.5 µs) 7.0 µs 11.5 µs 11.5 µs
mode() - n=500 25.3 µs 39,520 ( 22.0 µs … 121.1 µs) 25.0 µs 49.9 µs 51.1 µs
quantile() - n=500 6.4 µs 155,300 ( 6.3 µs … 7.2 µs) 6.4 µs 7.2 µs 7.2 µs
range() - n=500 358.6 ns 2,788,000 (339.6 ns … 379.9 ns) 366.3 ns 374.8 ns 379.9 ns
vari() - n=500 598.7 ns 1,670,000 (591.2 ns … 664.2 ns) 596.0 ns 664.2 ns 664.2 ns
std() - n=500 595.3 ns 1,680,000 (591.2 ns … 626.5 ns) 595.2 ns 626.5 ns 626.5 ns
cov() - n=500,2 905.5 ns 1,104,000 (895.5 ns … 962.2 ns) 904.6 ns 962.2 ns 962.2 ns
corr() - n=500,2 1.5 µs 673,900 ( 1.4 µs … 2.2 µs) 1.5 µs 2.2 µs 2.2 µs
kurtosis() - n=500 4.6 µs 219,400 ( 4.5 µs … 4.6 µs) 4.6 µs 4.6 µs 4.6 µs
skewness() - n=500 4.6 µs 219,000 ( 4.5 µs … 4.7 µs) 4.6 µs 4.7 µs 4.7 µs
ss() - n=500 637.8 ns 1,568,000 (590.7 ns … 2.5 µs) 625.7 ns 2.5 µs 2.5 µs
ssDiff() - n=500,2 330.6 ns 3,025,000 (312.1 ns … 344.0 ns) 336.2 ns 341.8 ns 344.0 ns
sem() - n=500 597.6 ns 1,673,000 (591.4 ns … 628.6 ns) 596.5 ns 628.6 ns 628.6 ns
Array.prototype.sort() - n=5000 1.3 ms 764.9 ( 1.2 ms … 3.1 ms) 1.3 ms 1.5 ms 1.8 ms
sort() - iterativeQuickSort - n=5000 121.1 µs 8,258 (106.5 µs … 291.8 µs) 124.3 µs 150.9 µs 159.1 µs
sort() - recursiveQuickSort - n=5000 149.1 µs 6,707 (130.8 µs … 294.5 µs) 152.8 µs 185.2 µs 205.8 µs
sort() - mergeSort - n=5000 340.3 µs 2,938 (302.5 µs … 470.2 µs) 344.7 µs 389.3 µs 396.0 µs
sort() - heapSort - n=5000 433.5 µs 2,307 (388.4 µs … 596.4 µs) 435.2 µs 504.7 µs 527.8 µs
z2p() 6.5 ns 153,800,000 ( 6.3 ns … 16.0 ns) 6.4 ns 9.8 ns 10.1 ns
p2z() 14.8 ns 67,350,000 ( 13.1 ns … 29.7 ns) 14.7 ns 19.8 ns 21.1 ns
t2p() - df=30 115.5 ns 8,659,000 (107.0 ns … 125.8 ns) 116.7 ns 124.5 ns 124.8 ns
p2t() - df=30 534.8 ns 1,870,000 (493.3 ns … 565.3 ns) 542.0 ns 558.4 ns 565.3 ns
f2p() - df=5,30 95.3 ns 10,490,000 ( 87.5 ns … 119.1 ns) 96.2 ns 105.2 ns 107.5 ns
p2f() - df=5,30 557.6 ns 1,793,000 (523.5 ns … 574.5 ns) 565.0 ns 574.5 ns 574.5 ns
c2p() - df=6 41.5 ns 24,080,000 ( 38.0 ns … 58.3 ns) 41.6 ns 49.3 ns 51.5 ns
p2c() - df=6 382.1 ns 2,617,000 (332.8 ns … 3.1 µs) 361.3 ns 820.7 ns 3.1 µs
randomNormal() 15.6 ns 64,280,000 ( 13.5 ns … 34.8 ns) 15.7 ns 20.1 ns 21.1 ns
randomT() - df=30 74.4 ns 13,440,000 ( 68.3 ns … 101.9 ns) 75.2 ns 83.8 ns 84.6 ns
randomF() - df=5,30 115.9 ns 8,630,000 (110.0 ns … 175.7 ns) 116.8 ns 135.0 ns 135.3 ns
randomChi2() - df=6 55.9 ns 17,900,000 ( 52.9 ns … 115.5 ns) 56.1 ns 63.9 ns 68.7 ns
KurtosisTest - n=500 4.8 µs 207,800 ( 4.6 µs … 8.2 µs) 4.7 µs 8.2 µs 8.2 µs
SkewnessTest - n=500 4.6 µs 215,600 ( 4.6 µs … 4.7 µs) 4.7 µs 4.7 µs 4.7 µs
OneWayAnova - n=500 6.6 µs 151,900 ( 6.0 µs … 128.6 µs) 6.5 µs 7.8 µs 10.5 µs
PeerAnova - n=500,3iv 18.5 µs 54,180 ( 17.3 µs … 150.9 µs) 18.3 µs 23.7 µs 41.4 µs
LeveneTest - n=500 21.3 µs 46,890 ( 20.5 µs … 154.2 µs) 21.1 µs 27.2 µs 32.0 µs
OneSampleKSTest - n=500 90.0 µs 11,110 ( 87.2 µs … 194.9 µs) 89.2 µs 139.0 µs 145.8 µs
OneSampleTTest - n=500 1.6 µs 626,100 ( 1.6 µs … 1.8 µs) 1.6 µs 1.8 µs 1.8 µs
TwoSampleTTest - n=500 3.0 µs 334,600 ( 2.9 µs … 4.3 µs) 3.0 µs 4.3 µs 4.3 µs
PeerSampleTTest - n=500 4.0 µs 250,700 ( 3.9 µs … 4.0 µs) 4.0 µs 4.0 µs 4.0 µs
WelchTTest - n=500 2.2 µs 453,900 ( 2.1 µs … 3.3 µs) 2.2 µs 3.3 µs 3.3 µs
PearsonCorrTest - n=500 1.8 µs 551,100 ( 1.8 µs … 1.9 µs) 1.8 µs 1.9 µs 1.9 µs
bootstrapSample() - n=500,B=5000 34.2 ms 29.3 ( 32.7 ms … 37.8 ms) 34.8 ms 37.8 ms 37.8 ms
bootstrapTest() - mean - n=500,B=5000 36.6 ms 27.3 ( 34.5 ms … 40.0 ms) 37.7 ms 40.0 ms 40.0 ms
bootstrapTest() - median - n=500,B=5000 127.0 ms 7.9 (125.9 ms … 130.1 ms) 127.2 ms 130.1 ms 130.1 ms
bootstrapTest() - ab - n=500,B=5000 78.3 ms 12.8 ( 75.4 ms … 87.0 ms) 78.1 ms 87.0 ms 87.0 ms
LinearRegressionOne - n=500 6.4 µs 156,900 ( 4.8 µs … 506.3 µs) 5.2 µs 14.0 µs 14.2 µs
LinearRegressionStandard - n=500,3iv 107.6 µs 9,298 ( 90.6 µs … 605.9 µs) 115.9 µs 160.3 µs 321.3 µs
LinearRegressionStepwise - n=500,3iv 829.4 µs 1,206 (738.6 µs … 1.3 ms) 853.6 µs 1.2 ms 1.2 ms
LinearRegressionSequential - n=500,3iv 311.7 µs 3,208 (268.0 µs … 635.8 µs) 314.2 µs 541.0 µs 572.5 µs
SimpleMediationModel - n=500 95.3 µs 10,490 ( 82.8 µs … 509.2 µs) 98.7 µs 116.2 µs 223.2 µs
SimpleMediationModel - n=500 - bootstrap (B=5000) 457.0 ms 2.2 (445.8 ms … 460.7 ms) 459.3 ms 460.7 ms 460.7 ms
AlphaRealiability - n=500,3iv 22.3 µs 44,770 ( 21.3 µs … 273.8 µs) 22.1 µs 24.9 µs 28.3 µs
Matrix.prototype.add() - 50x100 17.4 µs 57,360 ( 16.9 µs … 356.9 µs) 17.3 µs 19.6 µs 20.7 µs
Matrix.prototype.transpose() - 50x100 19.9 µs 50,250 ( 19.0 µs … 271.7 µs) 19.9 µs 22.0 µs 23.8 µs
Matrix.prototype.multiply() - 50x100 * 100x50 1.1 ms 936.9 ( 1.1 ms … 1.2 ms) 1.1 ms 1.1 ms 1.1 ms
Matrix.prototype.inverse() - 50x50 453.0 µs 2,208 (444.2 µs … 737.9 µs) 452.3 µs 473.9 µs 500.0 µs