Линейная модель.

Linear model: Y = a + b*X

Coefficients

	Least Squares	Standard	T
Parameter	Estimate	Error	Statistic	P-Value
Intercept	1517.72	638.711	2.37622	0.0195
Slope	-2.07708	2.12035	-0.979591	0.3298

Analysis of Variance

Source	Sum of Squares	Df	Mean Square	F-Ratio	P-Value
Model	3.05868E7	1	3.05868E7	0.96	0.3298
Residual	2.96433E9	93	3.18746E7
Total (Corr.)	2.99492E9	94

Correlation Coefficient = -0.101059

R-squared = 1.02129 percent

R-squared (adjusted for d.f.) = -0.0429978 percent

Standard Error of Est. = 5645.76

Mean absolute error = 2191.33

Durbin-Watson statistic = 0.232882 (P=0.0000)

Lag 1 residual autocorrelation = 0.532628

Col_2 = 1517.72 - 2.07708*Col_1

Since the P-value in the ANOVA table is greater or equal to 0.05, there is not a statistically significant relationship between Col_2 and Col_1 at the 95.0% or higher confidence level.

The R-Squared statistic indicates that the model as fitted explains 1.02129% of the variability in Col_2. The correlation coefficient equals -0.101059, indicating a relatively weak relationship between the variables. The standard error of the estimate shows the standard deviation of the residuals to be 5645.76. This value can be used to construct prediction limits for new observations by selecting the Forecasts option from the text menu.

The mean absolute error (MAE) of 2191.33 is the average value of the residuals. The Durbin-Watson (DW) statistic tests the residuals to determine if there is any significant correlation based on the order in which they occur in your data file. Since the P-value is less than 0.05, there is an indication of possible serial correlation at the 95.0% confidence level. Plot the residuals versus row order to see if there is any pattern that can be seen.

Мультипликативная модель.

Multiplicative model: Y = a*X^b

Coefficients

	Least Squares	Standard	T
Parameter	Estimate	Error	Statistic	P-Value
Intercept	9.81502	0.465532	21.0835	0.0000
Slope	-1.90126	0.113276	-16.7844	0.0000

NOTE: intercept = ln(a)

Analysis of Variance

Source	Sum of Squares	Df	Mean Square	F-Ratio	P-Value
Model	601.196	1	601.196	281.72	0.0000
Residual	198.466	93	2.13404
Total (Corr.)	799.662	94

Correlation Coefficient = -0.867071

R-squared = 75.1812 percent

R-squared (adjusted for d.f.) = 74.9144 percent

Standard Error of Est. = 1.46084

Mean absolute error = 1.21891

Durbin-Watson statistic = 0.125032 (P=0.0000)

Lag 1 residual autocorrelation = 0.885756

Col_2 = exp(9.81502 - 1.90126*ln(Col_1))

Unusual Residuals

			Predicted		Studentized
Row	X	Y	Y	Residual	Residual
93	1357.0	1.0	0.0202652	0.979735	2.89
94	1429.0	1.0	0.0183681	0.981632	2.97
95	1796.0	1.0	0.0118937	0.988106	3.35

Since the P-value in the ANOVA table is less than 0.05, there is a statistically significant relationship between Col_2 and Col_1 at the 95.0% confidence level.

The R-Squared statistic indicates that the model as fitted explains 75.1812% of the variability in Col_2. The correlation coefficient equals -0.867071, indicating a moderately strong relationship between the variables. The standard error of the estimate shows the standard deviation of the residuals to be 1.46084. This value can be used to construct prediction limits for new observations by selecting the Forecasts option from the text menu.

The mean absolute error (MAE) of 1.21891 is the average value of the residuals. The Durbin-Watson (DW) statistic tests the residuals to determine if there is any significant correlation based on the order in which they occur in your data file. Since the P-value is less than 0.05, there is an indication of possible serial correlation at the 95.0% confidence level. Plot the residuals versus row order to see if there is any pattern that can be seen.