Регрессионный анализ в пакетах STATGRAPHICS и MATHCAD, страница 2

-----------------------------------------------------------------------------90,00%                    90,00%

Predicted        Prediction Limits         Confidence Limits

X            Y        Lower        Upper        Lower        Upper

-----------------------------------------------------------------------------10,0      9,73359      7,84817       11,619      9,12052      10,3467

110,0     -95,9871     -111,814     -80,1599     -111,714     -80,2607

-----------------------------------------------------------------------------The StatAdvisor

--------------This table shows the predicted values for Prod using the fitted

model.  In addition to the best predictions, the table shows:

(1) 90,0% prediction intervals for new observations

(2) 90,0% confidence intervals for the mean of many observations

The prediction and confidence intervals correspond to the inner and

outer bounds on the graph of the fitted model.

Comparison of Alternative Models

-------------------------------------------------Model                  Correlation       R-Squared

-------------------------------------------------Linear                     -0,9795          95,94%

Square root-Y              -0,9759          95,24%

Square root-X              -0,9724          94,55%

Exponential                -0,9678          93,65%

Logarithmic-X              -0,9484          89,95%

Reciprocal-Y                0,9414          88,61%

Multiplicative             -0,9080          82,44%

Reciprocal-X                0,8558          73,25%

S-curve                     0,7932          62,91%

Double reciprocal          -0,7263          52,75%

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 linear model yields the

highest R-Squared value with 95,9444%.  This is the currently selected

model.

Unusual Residuals

---------------------------------------------------------------------------Predicted                 Studentized

Row                X             Y             Y      Residual      Residual

------------------------------------------------------------------------------------------------------------------------------------------------------The StatAdvisor

--------------The table of unusual residuals lists all observations which have

Studentized residuals greater than 2.0 in absolute value.  Studentized

residuals measure how many standard deviations each observed value of

Prod deviates from a model fitted using all of the data except that

observation.  In this case, there are no Studentized residuals greater

than 2.0. 

Influential Points

---------------------------------------------------------------------------Predicted   Studentized

Row                X             Y             Y      Residual      Leverage

------------------------------------------------------------------------------------------------------------------------------------------------------Average leverage of single data point = 0,222222

The StatAdvisor

--------------The table of influential data points lists all observations which

have leverage values greater than 3 times that of an average data

point.  Leverage is a statistic which measures how influential each

observation is in determining the coefficients of the estimated model.

In this case, an average data point would have a leverage value equal

to 0,222222.  There are no data points with more than 3 times the

average leverage. 

Simple Regression - Prod vs. Temp

MATHCAD: