Let x be the no. of mistakes is a poisson variate the parameter is obtained as below: Let f be the no. of pages.
From the above distribution:
Theoretical frequency: Tx = p{x) N
T0 =p(x = 0)100 x 100
= 0.3012 x 100 = 30.12
T5 and more = N – [T0 + T1 + T2 + T3 + T4
= 100 – [30 + 36 + 22 + 9 + 3]
= 100 – 100 = 0
The fitted observed and theoretical frequency distribution is:
chi – Square test: H0 : Poisson distribution is a good fit [Oi = Ei]
Hj: poisson distribution is not good fit [Oi = Ei] {Upper tail test}
Under H0, the χ2 – test stastic is:
λ is estimated from the parameter and so degrees of freedom is (n – 2)
Ay α = 5% for (n – 2) = 4.2 = 2 d.f the upper tail critical value k2= 5.99
Here χ2call lies in acceptance region (A.R)
.’. Ho is accepted.
Conclusion: Poisson distribution is good fit.