91影视

Let,

\({d_{ij}} = {x_i} - {y_i}\)

Where\({x_1},{x_2},....,{x_m}\)and\({y_1},{y_2},...{y_n}\)are observed values of continuous distribution that only in the location but not in shap the general form of a\(100(1 - \alpha )\)confidence interval for\({\mu _1} - {\mu _2}\)

Is,

\(\left( {{d_{ij\left( {mn - c + 1} \right)}},{d_{ij\left( c \right)}}} \right)\)

Where\({d_{ij\left( 1 \right)}},{d_{ij\left( 2 \right)}}...,{d_{ij\left( {mn} \right)}}\)are the ordered difference, use appendix values of c.

For m = 5, n = 5 and for 94.5% confidence interval, c = 22, the confidence interval is the form of

\(\left( {{d_{ij\left( {mn - c + 1} \right)}},{d_{ij\left( c \right)}}} \right) = \left( {{d_{ij\left( {5.5 - 22 + 1} \right)}},{d_{ij\left( {22} \right)}}} \right) = \left( {{d_{ij\left( 4 \right)}},{d_{ij\left( {22} \right)}}} \right)\)

Compute all difference, find 4^th and 23th ordered difference values. The four smallest ordered difference are

\( - 7.1, - 6.5, - 6.1. - 5.9\)

And four large ordered differences are

\( - 3.8, - 3.7, - 3.4. - 3.2\)

The 4^th and 22th ordered difference are -5.9 and -3.8, respectively; a 94.5% confidence interval is

\(\left( {{d_{ij\left( 4 \right)}},{d_{ij\left( {22} \right)}}} \right)\)=(-5.9, -3.8)