Добавил:
Upload Опубликованный материал нарушает ваши авторские права? Сообщите нам.
Вуз: Предмет: Файл:
Вязкость Стат.методы.doc
Скачиваний:
5
Добавлен:
12.09.2019
Размер:
60.93 Кб
Скачать

Analyze Experiment - нук_1

Analysis of Variance for НУК_1

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

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

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

A:H202:АсОН_A 12,587 1 12,587 44,92 0,0000

B:Продолжительность_ 8,7723 1 8,7723 31,31 0,0001

AA 3,52814 1 3,52814 12,59 0,0036

blocks 0,355606 1 0,355606 1,27 0,2803

Total error 3,64278 13 0,280214

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

Total (corr.) 28,8858 17

R-squared = 87,389 percent

R-squared (adjusted for d.f.) = 84,6867 percent

Standard Error of Est. = 0,529352

Mean absolute error = 0,334938

Durbin-Watson statistic = 1,59792 (P=0,1485)

Lag 1 residual autocorrelation = 0,188613

Cannot conduct lack-of-fit test.

No degrees of freedom for pure error.

The StatAdvisor

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

The ANOVA table partitions the variability in НУК_1 into separate

pieces for each of the effects. It then tests the statistical

significance of each effect by comparing the mean square against an

estimate of the experimental error. In this case, 3 effects have

P-values less than 0,05, indicating that they are significantly

different from zero at the 95,0% confidence level.

The lack of fit test is designed to determine whether the selected

model is adequate to describe the observed data, or whether a more

complicated model should be used. The test is performed by comparing

the variability of the current model residuals to the variability

between observations at replicate settings of the factors.

Unfortunately, the test can not be run in this case because there are

no replicate observations.

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

87,389% of the variability in НУК_1. The adjusted R-squared

statistic, which is more suitable for comparing models with different

numbers of independent variables, is 84,6867%. The standard error of

the estimate shows the standard deviation of the residuals to be

0,529352. The mean absolute error (MAE) of 0,334938 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 greater than 0.05, there is no indication of serial autocorrelation

in the residuals.

Regression coeffs. for НУК_1

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

constant = 5,89

A:H202:АсОН_A = 1,02417

B:Продолжительность_B = 0,855

AA = -0,939167

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

The StatAdvisor

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

This pane displays the regression equation which has been fitted to

the data. The equation of the fitted model is

НУК_1 = 5,89 + 1,02417*H202:АсОН_A + 0,855*Продолжительность_B -

0,939167*H202:АсОН_A^2

where the values of the variables are specified in their original

units. To have STATGRAPHICS evaluate this function, select

Predictions from the list of Tabular Options. To plot the function,

select Response Plots from the list of Graphical Options.

Estimation Results for НУК_1

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

Observed Fitted Lower 95,0% CL Upper 95,0% CL

Row Value Value for Mean for Mean

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

1 2,66 2,93111 2,29896 3,56326

2 4,56 4,89444 4,2623 5,52659

3 4,94 4,97944 4,3473 5,61159

4 3,8 3,78611 3,24701 4,32521

5 6,08 5,74944 5,21035 6,28854

6 5,07 5,83444 5,29535 6,37354

7 4,56 4,64111 4,00896 5,27326

8 7,22 6,60444 5,9723 7,23659

9 7,22 6,68944 6,0573 7,32159

10 3,04 3,21222 2,58007 3,84437

11 5,32 5,17556 4,54341 5,8077

12 5,7 5,26056 4,62841 5,8927

13 4,56 4,06722 3,52812 4,60632

14 6,46 6,03056 5,49146 6,56965

15 6,08 6,11556 5,57646 6,65465

16 4,94 4,92222 4,29007 5,55437

17 5,7 6,88556 6,25341 7,5177

18 6,84 6,97056 6,33841 7,6027

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

The StatAdvisor

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

This table contains information about values of НУК_1 generated

using the fitted model. The table includes:

(1) the observed value of НУК_1 (if any)

(2) the predicted value of НУК_1 using the fitted model

(3) 95,0% confidence limits for the mean response

Each item corresponds to the values of the experimental factors in a

specific row of your data file. To generate forecasts for additional

combinations of the factors, add additional rows to the bottom of your

data file. In each new row, enter values for the experimental factors

but leave the cell for the response empty. When you return to this

pane, forecasts will be added to the table for the new rows, but the

model will be unaffected.

Optimize Response

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

Goal: maximize НУК_1

Optimum value = 7,02421

Factor Low High Optimum

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

H202:АсОН_A -1,0 1,0 0,54517

Продолжительность_B -1,0 1,0 1,0

The StatAdvisor

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

This table shows the combination of factor levels which maximizes

НУК_1 over the indicated region. Use the Analysis Options dialog box

to indicate the region over which the optimization is to be performed.

You may set the value of one or more factors to a constant by setting

the low and high limits to that value.