SAS / PC
十五、 範 例 十五: 一般線性模式 (PROC GLM)
程式檔名稱 : SAMPLE15
| DATA RABBIT; | ||||||
| 原始資料輸入 | ||||||
| INPUT | GENETIC | SEX | AGE | LEAN; | ||
| CARDS; | GENETIC | 不同遺傳背景之組別 | ||||
| 1 | 1 | 75 | 1.7 | SEX | 性別 | |
| 1 | 2 | 75 | 1.8 | AGE | 日齡 | |
| 1 | 1 | 72 | 1.6 | LEAN | 瘦肉率 | |
| 1 | 2 | 72 | 1.8 | |||
| 1 | 2 | 70 | 1.9 | |||
| 1 | 2 | 90 | 1.9 | |||
| 2 | 1 | 95 | 1.1 | 註: 當雙向分類或雙向以上之變方分析如 | ||
| 2 | 1 | 74 | 1.2 | 樣品數不等時, 則宜採用 PROC GLM | ||
| 2 | 1 | 80 | 1.3 | |||
| 2 | 2 | 82 | 1.4 | PROC GLM 包含了 PROC ANOVA 之功 | ||
| 2 | 1 | 90 | 1.3 | 能, 在 N 相等時可選用 PROC ANOVA | ||
| 2 | 2 | 81 | 1.4 | 其運算速度較快, 但 PROC ANOVA 不 | ||
| 3 | 1 | 92 | 1.6 | 能置放共變數 (COVARIATE), 如此例 | ||
| 3 | 2 | 83 | 1.2 | 之 AGE | ||
| 3 | 1 | 83 | 1.1 | |||
| 3 | 2 | 84 | 1.3 | |||
| ; | ||||||
| RUN;
|
||||||
| PROC GLM; | ||||||
| CLASS GENETIC SEX; | 處理欄位等質量變數宣告 (CLASS) | |||||
| MODEL LEAN=GENETIC SEX GENETIC*SEX AGE; | 如 PROC ANOVA 之模式設定, | |||||
| 但 GLM 可放置共變數 (數量 | ||||||
| 變數) 為應變數, 即其可作共 | ||||||
| 變方分析 (表 15-1) | ||||||
| LSMEANS GENETIC SEX/PDIFF STDERR; | 最小平方平均值之計算、差異性 | |||||
| 檢定 (PDIFF), 及其標準機差計 | ||||||
| 算 (表 15-2) | ||||||
| CONTRAST 'GENETIC 1 VS 2 3' GENETIC -2 1 1; | 可用 CONTRAST 指令, 進行不同 | |||||
| RUN; | 設計之直交多項式比較 | |||||
| (表 15-3) | ||||||
|
表 15-1 |
||||||
|
SAS |
||||||
|
General Linear Models Procedure |
||||||
| Dependent Variable: LEAN | ||||||
|
Sum of Mean |
||||||
| Source DF Squares Square F Value Pr > F | ||||||
| Model 6 1.02551353 0.17091892 9.35 0.0019 | ||||||
| Error 9 0.16448647 0.01827627 | ||||||
| Corrected Total 15 1.19000000 | ||||||
| R-Square C.V. Root MSE LEAN Mean | ||||||
| 0.861776 9.165408 0.1351898 1.47500000 | ||||||
|
Source DF Type I SS Mean Square F Value Pr > F |
||||||
|
GENETIC |
2 | 0.91333333 | 0.45666667 | 24.99 | 0.0002 | |
| SEX | 1 | 0.04363636 | 0.04363636 | 2.39 | 0.1567 | |
| GENETIC*SEX | 2 | 0.06053030 | 0.03026515 | 1.66 | 0.2441 | |
| AGE | 1 | 0.00801353 | 0.00801353 | 0.44 | 0.5245 | |
| Source | DF | Type III SS | Mean Square | F Value | Pr > F | |
| GENETIC | 2 | 0.51892887 | 0.25946443 | 14.20 | 0.0016 | |
| SEX | 1 | 0.03328156 | 0.03328156 | 1.82 | 0.2102 | |
| GENETIC*SEX | 2 | 0.05292475 | 0.02646238 | 1.45 | 0.2850 | |
| AGE | 1 | 0.00801353 | 0.00801353 | 0.44 | 0.5245 | |
|
表 15-2 |
||||||
|
Least Squares Means |
||||||
| GENETIC LEAN Std Err Pr > |T| LSMEAN | ||||||
| LSMEAN LSMEAN H0:LSMEAN=0 Number | ||||||
| 1 |
1.77262644 |
0.06778206 | 0.0001 | 1 | ||
| 2 | 1.30495785 | 0.05963670 | 0.0001 | 2 | ||
| 3 | 1.28350155 | 0.07204072 | 0.0001 | 3 | ||
| Pr > |T| H0: LSMEAN(i)=LSMEAN(j) | ||||||
| i/j 1 2 3 | ||||||
| 1 . 0.0008 0.0014 | ||||||
| 2 0.0008 . 0.8178 | ||||||
| 3 0.0014 0.8178 . | ||||||
| NOTE: To ensure overall protection level, only probabilities | ||||||
| associated with pre-planned comparisons should be used. | ||||||
|
SEX LEAN Std Err Pr > |T| Pr > |T| H0: |
||||||
| LSMEAN LSMEAN H0:LSMEAN=0 LSMEAN1=LSMEAN2 | ||||||
| 1 1.40534790 0.05058358 0.0001 0.2102 | ||||||
| 2 1.50204266 0.05047660 0.0001 | ||||||
|
表 15-3 |
||||||
| Dependent Variable: LEAN | ||||||
| Contrast DF Contrast SS Mean Square F Value Pr > F | ||||||
| GENETIC 1 VS 2 3 1 0.51237564 0.51237564 28.04 0.0005 | ||||||