Quantitative – Regression (Multiple)

The Regression (Correlation & Multiple) VISA procedures allow the user to analyze problems which involve the
relationship between a Dependent Variable and up to 6 Independent Variables. Data is entered for up to 1000
observations for each in worksheet ED.
Seven analytical sections are provided:
• Scatter Plot & Coefficient of Correlation
• Correlation Table, Analyze Regression Output
• Analyze Residual Patterns
• Check Residuals for Normality
• Check Residuals for Autocorrelation
• Using the Regression Equation
The worksheets for these procedures are:
• GRAPH – allows the user to view an XY Scatter Plot showing the relationship between each Independent
variable and Dependent variable. The Coefficient of Correlation is also calculated for each with a display
of whether the relationship is statistically significant at the .05 level.
• COR – displays the Correlation Table showing the correlation between each Independent variable and
Dependent variable. It also shows the correlation between Independent variables and identifies “highly
correlated” (multi-co linear) relationships. It also identifies the “best” one Independent variable model.
• ARO – the user is provided an option box to select which Independent variables to include in the
regression model. If two “highly correlated” Independent variables are selected the worksheet warns the
user that “multi-co linearity” exists. The regression output is automatically calculated once the
Independent variables are selected. The user is immediately able to view the Regression: Summary Output
(Coefficient of Correlation, Coefficient of Determination, Adjusted R Square, and Standard Error);
ANOVA Table (p-value for Model Significance), Coefficient Table (Intercept, Slope coefficients and pvalue
for each Independent variable Significance); and the Regression Equation. The user enters a
significance level and the Model Significance and each Independent Variable Significance is automatically
displayed.
• ARES – displays the Predicted and Residual values with plots showing all possible relationships:
Residuals versus Predicted value; Residuals versus Observation sequence; and Residuals versus each
selected Independent variable.
• ARESN – allows the user to view a histogram of the Residuals compared to the expected Normal
distribution. Also, the worksheet calculates the Chi-Square test for Normality based on a user input
significance level.
• ARESA – provides the Durbin Watson Statistic for evaluation of Autocorrelation
• USING – the user can input a value for each Independent variable within the Relevant Range. The Point
Estimate of the Dependent value is automatically calculated. Selection of a confidence level allows the
user to view the Predication and Estimation Confidence intervals.

Tinggalkan Balasan

Please log in using one of these methods to post your comment:

Logo WordPress.com

You are commenting using your WordPress.com account. Logout / Ubah )

Gambar Twitter

You are commenting using your Twitter account. Logout / Ubah )

Foto Facebook

You are commenting using your Facebook account. Logout / Ubah )

Foto Google+

You are commenting using your Google+ account. Logout / Ubah )

Connecting to %s

%d blogger menyukai ini: