Contents

Examples

Data Size: Different versions of XLMiner™  have varying limits on size of data. The size of data depicted in the example below may not be supported by your version. Refer to Data Handling Specifications for details.

We will use the dataset Boxplot.xls for drawing the box plots of different specifications.

Open this dataset and try different combinations of specifications on it.

Select XLMiner --> Charts --> Box Plot.

  1. Select Y1 as Y-Variable . Keep the default option of "Notched" checked. Enter 0.99 for Confidence level. Change the Y-Axis title to "Y1 Values".

    Click OK. The output of Box plot will be shown.

    The box plot shows the title we have specified for Y1.

    We can calculate mean, Q1, Q2 using formulas. They are --

    Mean = 23.66

    Q1 = 10.49

    Q3 = 37.71

    The scale shown in the plot above confirms these values.

    Substituting these values ,

    Cutoff1 = 10.49 - 1.5 * (37.71 - 10.49) (Refer to the formula in the introduction.)

       = - 30.34

    Min = The actual value in the database which is just >= Cutoff1 = -19.32.

    The values like -47.343 and -32.343 are less than Min, so shown as outliers.

    The exercise above was just to illustrate the utility of the Box plot. We need not calculate the above values.  The box plot displays them to us once we specify the dataset. Read the scale and confirm the values. Then try to estimate the value of Max from the plot. Using the formula from introduction, find out Cutoff2 and verify if the Max you read from the plot is just <= Cutoff2.

  2. Select Y1 and Y2 as Y-Variables. Let the other default options stay.

    Let us see how the box plot for two variables appears.

    There are two plots. The plot indicates that the box in black is for the variable Y1 and the one in red is for variable Y2. The scale is common to both the plots. Try reading the Q1, Q2, Min, Max, Mean from the plot of Y2.

  3. Enter the following specifications. Note that we have introduced the X-Variable as well. The other defaults are not altered.
     

    Have a look at the output.

     

    The X-Var assumes two distinct values, (see the dataset shown in the beginning.), 3 and 4. XLMiner™ draws two plots for variable Y1, one each for each distinct value of X-Var. Same for Y2. So there are four boxes.

     

  4.  Let us see what happens if the option "Notched" is not checked.

     

We have chosen the X and Y variables as earlier but not checked "Notched". See the output.

 

The output resembles the one in example 3, except that the boxes are not notched. (The notch-height represents the confidence interval around the mean. When we de-check "Notched" we do not expect the confidence interval to be displayed.)

 

Note : On invoking the Charts menu, you will see an option "Delete hidden sheets". When XLMiner™ creats charts, some hidden sheets are created for internal calculations. When you delete the charts the hidden sheets are not deleted automatically. So when you delete the charts, select  "Delete hidden sheets"  once to delete the hidden sheets.

See also: