The probability integral of the correlation coefficient in samples from a normal bivariate population. (Q558854)
From MaRDI portal
| This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: The probability integral of the correlation coefficient in samples from a normal bivariate population. |
scientific article; zbMATH DE number 2546726
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | The probability integral of the correlation coefficient in samples from a normal bivariate population. |
scientific article; zbMATH DE number 2546726 |
Statements
The probability integral of the correlation coefficient in samples from a normal bivariate population. (English)
0 references
1933
0 references
Aus einer normal verteilten bivariaten Bevölkerung, in der der Korrelationskoeffizient gleich \(\varrho \) ist, mögen Stichproben der Größenordnung \(n\) herausgegriffen werden. Nach \textit{R. A. Fisher} ist die Wahrscheinlichkeit dafür, daß\ der Korrelationskoeffizient der Stichprobe zwischen \(r\) und \(r+dr\) liegt, gleich \(y_n(r)dr\), wobei \[ y_n(r)=\frac {(1-\varrho ^2)^{\tfrac {n-1}{2}}}{(n-3)!\pi } (1-r^2)^{\tfrac {n-4}{2}} \frac {d^{n-2}}{d(r\varrho )^{n-2}} \left [ \frac {\cos ^{-1}(-\varrho r)}{\sqrt {1-\varrho ^2r^2}}\right ]. \] Verf. berechnet das Integral \[ \int \limits _{-1}^{r}y_n(r)dr. \]
0 references