* A Poisson Model for Count Data * Reference: W.H. Greene, Econometric Analysis. * Second Edition - Section 21.8, pp. 676-8 * Third Edition - Section 19.9, pp. 931-4 * Fourth Edition - Section 19.9, pp. 880-2 * SAMPLE 1 20 * Data on ship accidents (TABLE 21.12 - Second Edition) * * Ship Type Year constructed Amount of * 60-4 65-9 70-4 75-9 Service Accidents * ------------- ------------------- READ OBS A B C D E YR60 YR65 YR70 YR75 SERV ACTUAL 1 1 0 0 0 0 1 0 0 0 127 0 2 1 0 0 0 0 0 1 0 0 63 4 3 1 0 0 0 0 0 0 1 0 1095 18 4 1 0 0 0 0 0 0 0 1 2244 11 5 0 1 0 0 0 1 0 0 0 17176 29 6 0 1 0 0 0 0 1 0 0 20370 53 7 0 1 0 0 0 0 0 1 0 13099 44 8 0 1 0 0 0 0 0 0 1 7117 18 9 0 0 1 0 0 1 0 0 0 1179 1 10 0 0 1 0 0 0 1 0 0 676 1 11 0 0 1 0 0 0 0 1 0 1948 2 12 0 0 1 0 0 0 0 0 1 274 1 13 0 0 0 1 0 1 0 0 0 251 0 14 0 0 0 1 0 0 1 0 0 288 0 15 0 0 0 1 0 0 0 1 0 1208 11 16 0 0 0 1 0 0 0 0 1 2051 4 17 0 0 0 0 1 1 0 0 0 45 0 18 0 0 0 0 1 0 1 0 0 789 7 19 0 0 0 0 1 0 0 1 0 2161 12 20 0 0 0 0 1 0 0 0 1 542 1 * Poisson Regression with the MLE command. * See the estimation results in TABLE 21.13 - Greene, Second Edition. * An unrestricted model MLE ACTUAL B C D E YR65 YR70 YR75 SERV / TYPE=EPOISSON GEN1 LLFUN=$LLF * A restricted model MLE ACTUAL B C D E SERV / TYPE=EPOISSON GEN1 LLFREST=$LLF * A likelihood ratio test GEN1 LRSTAT=2*(LLFUN-LLFREST) PRINT LRSTAT STOP