, STATGRAPHICS.
1. Simple Regression.
1.1. Regression Analysis Linear model.
:
Dependent variable: Independent variable: | ||||
Parameter | Estimate | Standard Error | T - Statistic | P-value |
Intercept Slope | ||||
Slope |
. , . , . , Parameter , . α (Intercept) β (Slope) y = α + β x. a b . a b Estimate. Standard Error a b. (4.17). , ( a, b , b 0 b 1). T - Statistic (4.18), (4.19), β 0 β 1 0. , : β 0 = 0 : β 1 = 0. , , . , , , . β 1 ( ). ϒ P- (P-value) - ( ). , P- , . , P- 0,05.
1.2. Analysis of Variance.
:
Source | Sum of Squares | Df | Mean Squares | F-ratio | P-value |
Model Residual | |||||
Total (Corr.) |
( : ) .
, (Sum of Squares) , Source : Model, Residual () Total (Corr.). , S(yi ŷi)2, , , . (3.3) Model + Residual = Total (Corr.).
|
|
Df 볻, : 1, n 2 n 1 . 4.3.6, 4.3.8 4.3.7.
Mean Squares , : Model / 1 Residual / (n 2).
ϒ (F-ratio) (F -) .
(P-value) P- F -, (. 4.3.8).
1.3.
, Analysis of Variance, . .
Correlation Coefficient - (3.11).
R - Squared ( ) R 2, (3.5). , ( )
(Model / Total (Corr.)) × 100 %.
Standard Error of Est. . S 2, (4.15). ,
Standard Error of Est. = (Residual Mean Square)1 / 2.
1.4. Durbin Watson statistic P- . . ó ( ) P-.
1.5. Lag 1 residual autocorrelation .
2. , STATGRAPHICS, , , .