Laborathori ya Rulete · Laborathori · Zwiṱatisiṱiki · minithi dza 8

Ndingo ya Chi-Square

Ndingo ya tshiṱandadi ya “zwivhalo izwi zwa 37 ndi zwine vhili ḽi songo kheluwaho ḽa nga zwi bveledza naa?”

Phindulo pfufhi

χ² = Σ (zwo vhonwaho − zwo lavhelelwaho)² / zwo lavhelelwaho kha mabuli oṱhe a 37. Vhambedzani na phaḓaladzo ya chi-square kha zwiimo zwa mbofholowo zwa 36 uri ni wane ndeme ya p.

tshivhalo tsha ndingo
Σ (O−E)²/E
df
36
mukano kha 5%
51.0
mukano kha 1%
58.6

U i vhalela

Tshivhalo tshi lavhelelwaho kha bulizwishumiswaBuliḼiṅwe ḽa mabuli a 37 o ṅwalwaho nomboro kha rotara. Ḽi dovha ḽa vhidzwa: slot, cup.Vhalani zwoṱhe → ḽiṅwe na ḽiṅwe ndi n/37. Kha buli ḽiṅwe na ḽiṅwe, andisani phambano vhukati ha zwo vhonwaho na zwo lavhelelwaho nga yone ine, ni kovhe nga zwo lavhelelwaho; ṱanganyani zwipiḓa zwa 37. Vhili ḽi songo kheluwaho ḽi ṋea ndeme dzi re tsini na 36 nga tshivhalo tsha vhukati (zwiimo zwa mbofholowolaborathoriZwiimo zwa mbofholowoTshivhalo tsha ndeme dzi ḓiimisaho kha tshiṱatisiṱiki. Kha ndingo ya chi-square kha mabuli a 37 ndi 36, ngauri zwivhalo zwi fanela u ṱanganela zwa ita tshivhalo tshoṱhe.Vhalani zwoṱhe →), nga phambuwo ya tshiṱandadimathematikiPhambuwo ya tshiṱandadiMudzi wa tshikwea wa vhariantsi; tshikhala tsha ḓuvha ḽiṅwe na ḽiṅwe vhukati ha mvelelo na ndeme yo lavhelelwaho.Vhalani zwoṱhe → ya 8.5. Nṱha ha 51.0 zwi na ndeme kha vhuimo ha 5%; nṱha ha 58.6 kha 1%. Tshisaukanyi tsha U Kheluwa ha Vhili tshi vhalela tshivhalo itsho na ndeme ya plaborathoriNdeme ya pKhonadzeo ya u vhona u kheluwa hu linganaho kana hu fhiraho hune ha vhonwa arali vhili ḽo vha ḽi songo kheluwa.Vhalani zwoṱhe → yo teaho nga fankishini ya gamma i songo fhelelaho yo lugiswaho.

mimono ya nzwi lavhelelwaho kha buliσ kha bulibuli kha +2σbuli kha +3σ
370103.11619
1,000275.13742
3,7001009.9120130
10,00027016.2303319
37,0001,00031.21,0621,094
σ ≈ √(n·p·(1−p)) hu na p = 1/37. Kha mimono ya 1,000 buli ḽo bvaho lwa 37 a ḽi na tshithu tsho khethekanaho; ḽo bvaho lwa 42 ḽi fanela tsumbo ya vhuvhili.

Tsumbo yo shumiwaho

Consider a hypothetical sample of 1,000 independent spinsmathematikiMimono i ḓiimisahoMumono muṅwe na muṅwe u na khonadzeo dzi fanaho hu sa londwi zwo fhiraho. Ndi mutheo wa u vhalela hoṱhe kha saiti ino.Vhalani zwoṱhe →. Each pocket has an expected count of 1,000/37 ≈ 27.03. If pocket 8 appears 41 times, its contribution to the statistic is about 7.22. Suppose the total statistic is approximately 45.2. On 36 degrees of freedom, the approximate p-value is 0.14: under the fair-wheel model, a statistic at least this large has a probabilitymathematikiKhonadzeoKhonadzeo ya mvelelo, u bva kha 0 u ya kha 1 kana 0% u ya kha 100%. Nomboro nthihi: 1/37 = 2.70%.Vhalani zwoṱhe → of about 14%. This is not significant at the 5% level, and it does not prove the wheel is fair. A pocket selected after examining these counts needs independent confirmation; another 1,000 spins is not guaranteed to resolve a small departure.

Zwine zwa dzhiiwa

The usual approximation assumes independent observations and adequate expected counts. The common five-per-cell rule gives n ≥ 185 for 37 equal pockets; it is a guideline for the approximation, not a hard boundary for all possible tests. Pooling wheels changes the population being tested. Dependence may invalidate the reference distribution without necessarily increasing the statistic.

Tsaino ya Mukurupiye wa RuleteMimono ya Rulete i Ḓiimisaho

Pearson na Monte Carlo

Karl Pearson o gandisa ndingo nga 1900, nahone iṅwe ya tsumbo dzawe dza u thoma yo vha i mimono ya 26,306 ya rulete ya Monte Carlo yo ṅwaliwaho, ye a i wana i sa lingani nga maanḓa lwa uri i sa vhe ya khonadzeo, a ri “rulete ya Monte Carlo… a si mutambo wa khonadzeo”. Vhaṅwali vha nga murahu vho wana uri vhoraḓiṅwalwa vhe vha ṅwala mimono iyo kha gurannda vho ṱumbula minzhi yayo, zwine zwa vha ngudo nga yoṱhe nga ha uri data i bva ngafhi.

Ndingo ya Pearson ya 1900 yo wana mimono ya Monte Carlo i sa lingani nga maanḓa; vhaṅwali vha gurannda vho vha vho ṱumbula minzhi yayo.

Rulete ya Monte Carlo

Ndingo ya vhukuma: mavhili a ḓishumaho ngei Las Vegas

A published analysis of specific automatic wheels is evidence about those recorded samples, not all automatic roulette. A non-significant uniformity test cannot prove the absence of steering, and combining several wheels does not identify the behavior of each one. Check the original study’s data and design before drawing that conclusion.

Nṱha ha chi-square

Kha buli ḽithihi ḽi timatimiswaho, ndingo ya bainomiala i na maanḓa u fhira. Kha tshipiḓa tshi timatimiswaho, ndingo ya z ya tshipiḓavhiliTshipiḓaMabuli a tevhekanaho kha vhili, a ita tshipiḓa tsha tshiṱuluwa, zwi fhambanaho na tshigwada kha mapa wa tafula.Vhalani zwoṱhe → kha siaṱari ḽa Tsenguluso ya zwipiḓa. Chi-square ndi ndingo ya zwoṱhe: i ni vhudza uri hu na tshi songo lugaho i sa ambi uri ndi mini. I shumiseni u thoma, nga murahu ni sedze.

Tsenguluso ya Zwipiḓa

Fomula

χ² = Σᵢ (Oᵢ − Eᵢ)² / Eᵢ
Eᵢ = n / 37 · df = 36
ndeme dza mukano: 51.0 (5%) · 58.6 (1%)

Mbudziso

Ndi ngani zwiimo zwa mbofholowo zwi 36 hu si 37?

Zwivhalo zwa 37 zwi fanela u ṱanganyiwa zwa vha n, ngauralo ndi zwa 36 fhedzi zwi re na mbofholowo ya u shanduka.

Vhili ḽo kheluwaho nga maanḓa ḽi nga sumbedza χ² ifhio?

Vhili ḽi re na buli ḽithihi ḽa 1/25 kha mimono ya 10,000 ḽi nga ṋea tshipiḓa tsha buli tshi re tsini na 60 tshoṱhe; χ² i tsini na 100, p i tsini na ziro nga vhukuma.

Ndi nga i ita kha mimono ya 200 naa?

Nga thekhiniki ee, hu tshi lavhelelwa 5.4 kha buli ḽiṅwe na ḽiṅwe. A i nga wani tshithu, nga nnḓa ha vhili ḽi re na buli ḽo valwaho nga gulu.

Zwa u sea

For 36 degrees of freedom, the approximate 5% upper-tail cutoff is 51.0.

For 36 degrees of freedom, the approximate 5% upper-tail cutoff is 51.0.

Zwiko

  1. Pearson, K. “On the criterion that a given system of deviations from the probable…” https://www.itl.nist.gov/div898/software/dataplot/refman1/auxillar/chsqgood.htm
  2. Feller, W. An Introduction to Probability Theory and Its Applications, Vol. 1. https://www.amstat.org/asa/files/pdfs/P-ValueStatement.pdf
tsumbo: ya Yurophu, ziro nthihimaitele ↗zwiko ↗