ADM 2304 Lecture Notes - Lecture 21: Analysis Of Variance, Total Variation, Test Statistic
Document Summary
Divide total deviation of y from y into two parts: yi y = yi y yi yi. Square both sides and sum over all observations: yi y. )2 i=1 n i=1 yi yi (cross products = 0) Sstotal = i (yi y )2 = (n 1)sy. Ssr = i yi y )2 = xi yi . If we let r be the sample correlation coef cient and be the population correlation coef cient, then testing is equivalent to testing . 1 r 2 n 2 r n 2. Multiple independent predictor (or explanatory) variables x1, , xk. Mark based on age, gender and number of study per week. Waiting time based on type of surgery, hospital and time of the year hours. Two-way table, matrix, array, spreadsheet or data set. 2 n y1 x11 x1k y2 x21 x2k yn xn1 xnk.