STAT231 Lecture : chapter7_chisquare.pdf
Document Summary
Example: smoking and high risk of heart disease. Example: smoking and high risk of heart disease: hypothesis: h0: smoking and high risk are independent, h1: smoking and high risk are not independent, to test this hypothesis we apply the same general principle: Example: smoking/ high risk of heart disease: a two-way table is often called cross-tabs (cross tabulations). 100: these are the observed sample frequencies. Example: smoking/ high risk of heart disease: let x be smoking status, let y be high risk, if x and y are independent, then the cell counts can be computed as follows: n* p(y,x)= n *p(y) p(x) Computations in a two-way table: p(y) and p(x) can be computed from the marginal frequencies. Example: smoking/ high risk of heart disease: observed. 1: where the sum is over all r*c cells of the table. Example: smoking/ high risk of heart disease: we sum over the 4 cells, we compare this value with the critical value of a.