Independent variables. The equation of the fitted model is

LOG(P) = 5,92527 + 0,00796706*time*LOG(C) - 0,198376*LOG(LAG(M;1)) -

0,512978*LOG(W)

Since the P-value in the ANOVA table is less than 0.01, there is a

statistically significant relationship between the variables at the

99% Confidence level.

The R-Squared statistic indicates that the model as fitted

explains 99,9282% of the variability in LOG(P). The adjusted

R-squared statistic, which is more suitable for comparing models with

different numbers of independent variables, is 99,9184%. The standard

error of the estimate shows the standard deviation of the residuals to

be 0,0130357. This value can be used to construct prediction limits

for new observations by selecting the Reports option from the text

menu. The mean absolute error (MAE) of 0,0102107 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 DW

value is less than 1.4, there may be some indication of serial

correlation. Plot the residuals versus row order to see if there is

any pattern which can be seen.

In determining whether the model can be simplified, notice that the

highest P-value on the independent variables is 0,0025, belonging to

LOG(LAG(M;1)). Since the P-value is less than 0.01, the highest order

term is statistically significant at the 99% confidence level.

Consequently, you probably don't want to remove any variables from the

model.

Comparison of Alternative Models G

--------------------------------------------------

Model Correlation R-Squared

--------------------------------------------------

Exponential 0,9911 98,23%

Square root-Y 0,9900 98,02%

Linear 0,9862 97,26%

Reciprocal-Y -0,9844 96,91%

Square root-X 0,9572 91,63%

Multiplicative 0,9224 85,07%

Logarithmic-X 0,8828 77,93%

Double reciprocal 0,7292 53,18%

S-curve -0,6698 44,86%

Reciprocal-X -0,6101 37,23%

Logistic <no fit>

Log probit <no fit>

--------------------------------------------------

The StatAdvisor

---------------

This table shows the results of fitting several curvilinear models

to the data. Of the models fitted, the exponential model yields the

highest R-Squared value with 98,2254%. This is the currently selected

model.

Regression Analysis - Exponential model: Y = exp(a + b*X)

-----------------------------------------------------------------------------

Dependent variable: G

Independent variable: time

-----------------------------------------------------------------------------

Standard T

Parameter Estimate Error Statistic P-Value

-----------------------------------------------------------------------------

Intercept 6,12169 0,0115294 530,963 0,0000

Slope 0,0255672 0,000649437 39,3682 0,0000

-----------------------------------------------------------------------------

Analysis of Variance

-----------------------------------------------------------------------------

Source Sum of Squares Df Mean Square F-Ratio P-Value

-----------------------------------------------------------------------------

Model 1,46915 1 1,46915 1549,86 0,0000

Residual 0,0265419 28 0,000947924

-----------------------------------------------------------------------------

Total (Corr.) 1,49569 29

Correlation Coefficient = 0,991087

R-squared = 98,2254 percent

Standard Error of Est. = 0,0307884

The StatAdvisor

---------------

The output shows the results of fitting a exponential model to

describe the relationship between G and time. The equation of the

fitted model is

G = exp(6,12169 + 0,0255672*time)

Since the P-value in the ANOVA table is less than 0.01, there is a

statistically significant relationship between G and time at the 99%

confidence level.

The R-Squared statistic indicates that the model as fitted explains

98,2254% of the variability in G after transforming to a logarithmic

scale to linearize the model. The correlation coefficient equals

0,991087, indicating a relatively strong relationship between the

variables. The standard error of the estimate shows the standard

deviation of the residuals to be 0,0307884. This value can be used to

construct prediction limits for new observations by selecting the

Forecasts option from the text menu.

Comparison of Alternative Models Tr

--------------------------------------------------

Model Correlation R-Squared

--------------------------------------------------

Square root-X -0,9513 90,50%

Logarithmic-X -0,9470 89,68%

Log probit -0,9457 89,43%

Multiplicative -0,9445 89,21%

Reciprocal-Y 0,9338 87,20%

Exponential -0,9322 86,90%

Logistic -0,9316 86,80%

Square root-Y -0,9313 86,73%

Linear -0,9304 86,56%

Reciprocal-X 0,8341 69,57%

S-curve 0,8276 68,50%

Double reciprocal -0,8211 67,42%

--------------------------------------------------

The StatAdvisor

---------------

This table shows the results of fitting several curvilinear models

to the data. Of the models fitted, the square root-X model yields the

highest R-Squared value with 90,5036%. This is 3,94354% higher than

the currently selected linear model. To change models, select the

Analysis Options dialog box.

Regression Analysis - Square root-X model: Y = a + b*sqrt(X)

-----------------------------------------------------------------------------

Dependent variable: Tr

Independent variable: time

-----------------------------------------------------------------------------

Standard T

Parameter Estimate Error Statistic P-Value

-----------------------------------------------------------------------------

Intercept 0,256258 0,00145272 176,399 0,0000

Slope -0,00582581 0,000363179 -16,0412 0,0000

-----------------------------------------------------------------------------

Analysis of Variance

-----------------------------------------------------------------------------

Source Sum of Squares Df Mean Square F-Ratio P-Value

-----------------------------------------------------------------------------

Model 0,00130681 1 0,00130681 257,32 0,0000

Residual 0,000137122 270,00000507858

-----------------------------------------------------------------------------

Total (Corr.) 0,00144393 28

Correlation Coefficient = -0,951334

R-squared = 90,5036 percent

Standard Error of Est. = 0,00225357

The StatAdvisor

---------------

The output shows the results of fitting a square root-X model to

describe the relationship between Tr and time. The equation of the

fitted model is

Tr = 0,256258 - 0,00582581*sqrt(time)

Since the P-value in the ANOVA table is less than 0.01, there is a

statistically significant relationship between Tr and time at the 99%

confidence level.

The R-Squared statistic indicates that the model as fitted explains

90,5036% of the variability in Tr. The correlation coefficient equals

-0,951334, indicating a relatively strong relationship between the

variables. The standard error of the estimate shows the standard

deviation of the residuals to be 0,00225357. This value can be used

to construct prediction limits for new observations by selecting the

Forecasts option from the text menu.

Summary Statistics for Irr

Count = 30

Average = 5,59333

Variance = 17,3597

Standard deviation = 4,16649

Minimum = -1,21

Maximum = 17,18

Stnd. skewness = 1,52864

Stnd. kurtosis = 0,631698

Sum = 167,8

The StatAdvisor

---------------

This table shows summary statistics for Irr. It includes measures

of central tendency, measures of variability, and measures of shape.

Of particular interest here are the standardized skewness and

standardized kurtosis, which can be used to determine whether the

sample comes from a normal distribution. Values of these statistics

outside the range of -2 to +2 indicate significant departures from

normality, which would tend to invalidate any statistical test

regarding the standard deviation. In this case, the standardized

skewness value is within the range expected for data from a normal

distribution. The standardized kurtosis value is within the range

expected for data from a normal distribution.

Multiple Regression Analysis 2MNK1 Y

-----------------------------------------------------------------------------

Dependent variable: Y

-----------------------------------------------------------------------------

Standard T

Parameter Estimate Error Statistic P-Value

-----------------------------------------------------------------------------

Irr 23,4205 7,70289 3,04048 0,0058

Yl 0,544344 0,171321 3,17733 0,0042

Kl -0,280307 0,133867 -2,09392 0,0475

Ll 20,4968 4,79022 4,27888 0,0003

Pl 15,7494 3,59437 4,38168 0,0002

-----------------------------------------------------------------------------

Analysis of Variance

-----------------------------------------------------------------------------

Source Sum of Squares Df Mean Square F-Ratio P-Value

-----------------------------------------------------------------------------

Model 3,71554E8 5 7,43109E7 4617,65 0,0000

Residual 370135,0 23 16092,8

-----------------------------------------------------------------------------

Total 3,71925E8 28

R-squared = 99,9005 percent

R-squared (adjusted for d.f.) = 99,8832 percent

Standard Error of Est. = 126,857

Mean absolute error = 90,5341

Durbin-Watson statistic = 2,41492

The StatAdvisor

---------------

The output shows the results of fitting a multiple linear

regression model to describe the relationship between Y and 5