10-3 Prediction Intervals and Variation 549 Prediction Intervals Access tech supplements, videos, and data sets at www.TriolaStats.com TECH CENTER Statdisk Statdisk provides the intercept and slope of the regression equation, the standard error of estimate (labeled “Standard Error”), and the coefficient of determination. These results are helpful in finding a prediction interval, but the actual prediction interval is not provided. 1. Click Analysis in the top menu. 2. Select Correlation and Regression from the dropdown menu. 3. Enter the desired significance level and select the two columns to be evaluated. 4. Click Evaluate. StatCrunch 1. Click Stat in the top menu. 2. Select Regression from the dropdown menu, then select Simple Linear from the submenu. 3. Select the columns to be used for the x variable and y variable. 4. For Prediction of Y enter the desired x value(s) and significance level. 5. Click Compute! Minitab 1. Complete the Minitab Regression procedure from Section 10-2 to get the regression equation. Minitab will automatically use this equation in this procedure. 2. Click Stat in the top menu. 3. Select Regression from the dropdown menu and select Regression—Predict from the submenu. 4. Select Enter individual values from the dropdown menu. 5. Enter the desired value(s) for the x variable. 6. Click the Options button and change the confidence level to the desired value. 7. Click OK twice. TI-83, 84 Plus Calculator TI-83>84 Plus results include the intercept (a) and slope of the regression equation (b), the standard error of estimate (s), and the coefficient of determination 1r22. These results are helpful in finding a prediction interval, but the actual prediction interval is not provided. 1. Press K, then select TESTS in the top menu. 2. Select LinRegTTest in the menu and press [. 3. Enter the list names for the x and y variables. Enter 1 for Freq and for b & r select 3 0 to test the null hypothesis of no correlation. 4. Select Calculate and press [ to view results. R R command: predict(lm(y~x), interval=“confidence”, conf.level=0.95) TIP: Results provided for each value of x and in same order that x values occur in data set. A complete list of R statistical commands is available at TriolaStats.com Excel XLSTAT Add-In 1. Enter the sample data in columns of the worksheet. 2. Enter the desired value(s) for x to be used for the prediction interval in a column. 3. Click on the XLSTAT tab in the Ribbon and then click Modeling data. 4. Select Linear regression from the dropdown menu. 5. Enter the range of cells containing the Y>Dependent variable data and X>Explanatory variable data. Check the Quantitative box under X>Explanatory variable. If the first data row includes a label, check the Variable labels box. 6. Click the Options tab and enter the desired confidence interval, such as 95. 7. Click the Prediction tab. 8. Check the Prediction box and in the Quantitative box enter the cell range containing the desired value(s) of x from Step 2. The first cell in the range must contain a value, not a label. 9. Click OK. The prediction interval(s) are in the Predictions for the new observations table. Excel (Data Analysis Add-In) Excel provides the intercept and slope of the regression equation, the standard error of estimate se (labeled “Standard Error”), and the coefficient of determination (labeled “R Square”). These results are helpful in finding a prediction interval, but the actual prediction interval is not provided. 1. Click on the Data tab in the Ribbon and then click the Data Analysis tab. 2. Select Regression under Analysis Tools and click OK. 3. For Input Y Range enter the data range for the dependent y variable. For Input X Range enter the data range for the independent x variable. 4. Check the Labels box if the first row contains a label. 5. Click OK to display the results.
RkJQdWJsaXNoZXIy NjM5ODQ=