MTB > let c4 = c3/c2 MTB > let c5 = loge(c4/(1 - c4)) MTB > print c1-c5 ROW C1 C2 C3 C4 C5 1 5 200 32 0.160 -1.65823 2 10 200 51 0.255 -1.07212 3 15 200 70 0.350 -0.61904 4 20 200 103 0.515 0.06002 5 30 200 148 0.740 1.04597 MTB > let c6 = c2*c4*(1 - c4) MTB > print c1-c6 ROW C1 C2 C3 C4 C5 C6 1 5 200 32 0.160 -1.65823 26.880 2 10 200 51 0.255 -1.07212 37.995 3 15 200 70 0.350 -0.61904 45.500 4 20 200 103 0.515 0.06002 49.955 5 30 200 148 0.740 1.04597 38.480 MTB > plot c5 c1 - 1.0+ * - C5 - - - 0.0+ * - - - * - -1.0+ * - - - * - -2.0+ ----+---------+---------+---------+---------+---------+--C1 5.0 10.0 15.0 20.0 25.0 30.0 MTB > regress c5 1 c1 c11 c12 The regression equation is C5 = - 2.19 + 0.109 C1 Predictor Coef Stdev t-ratio p Constant -2.18602 0.05722 -38.20 0.000 C1 0.108584 0.003150 34.47 0.000 s = 0.06059 R-sq = 99.7% R-sq(adj) = 99.7% Analysis of Variance SOURCE DF SS MS F p Regression 1 4.3624 4.3624 1188.27 0.000 Error 3 0.0110 0.0037 Total 4 4.3735 MTB > regress c5 1 c1 c21 c22; SUBC> weights c6. The regression equation is C5 = - 2.19 + 0.109 C1 Predictor Coef Stdev t-ratio p Constant -2.18506 0.06783 -32.21 0.000 C1 0.108700 0.003632 29.93 0.000 Analysis of Variance SOURCE DF SS MS F p Regression 1 151.98 151.98 895.53 0.000 Error 3 0.51 0.17 Total 4 152.49 MTB > print c1 c4 c12 c22 ROW C1 C4 C12 C22 1 5 0.160 -1.64310 -1.64156 2 10 0.255 -1.10018 -1.09806 3 15 0.350 -0.55726 -0.55455 4 20 0.515 -0.01435 -0.01105 5 30 0.740 1.07149 1.07595 MTB > let c13 = expo(c12)/(1 + expo(c12)) MTB > let c23 = expo(c22)/(1 + expo(c22)) MTB > print c1 c4 c13 c23 ROW C1 C4 C13 C23 1 5 0.160 0.162044 0.162253 2 10 0.255 0.249706 0.250104 3 15 0.350 0.364181 0.364808 4 20 0.515 0.496414 0.497237 5 30 0.740 0.744880 0.745727 =================================================== Worksheet size: 40000000 cells MTB > help blog MTB > help blog MTB > blog c3 c2 = c1; SUBC> ST. Binary Logistic Regression Link Function: Logit Response Information Variable Value Count C3 Success 404 Failure 596 C2 Total 1000 Logistic Regression Table Odds 95% CI Predictor Coef StDev Z P Ratio Lower Upper Constant -2.1855 0.1647 -13.27 0.000 C1 0.108719 0.008843 12.29 0.000 1.11 1.10 1.13 Log-Likelihood = -584.382 Test that all slopes are zero: G = 180.437, DF = 1, P-Value = 0.000 Goodness-of-Fit Tests Method Chi-Square DF P Pearson 0.509 3 0.917 Deviance 0.510 3 0.917 Hosmer-Lemeshow 0.509 3 0.917 Table of Observed and Expected Frequencies: (See Hosmer-Lemeshow Test for the Pearson Chi-Square Statistic) Group Value 1 2 3 4 5 Total Success Obs 32 51 70 103 148 404 Exp 32.4 50.0 73.0 99.4 149.1 Failure Obs 168 149 130 97 52 596 Exp 167.6 150.0 127.0 100.6 50.9 Total 200 200 200 200 200 1000 Measures of Association: (Between the Response Variable and Predicted Probabilities) Pairs Number Percent Summary Measures Concordant 157311 65.3% Somers' D 0.47 Discordant 43711 18.2% Goodman-Kruskal Gamma 0.57 Ties 39762 16.5% Kendall's Tau-a 0.23 Total 240784 100.0% MTB >