Contents 
 
The Column Info Window
Use the Column Info window to specify all of the attributes and properties of a column.
To access the Column Info window:
• Select Cols > New Column
or
• Right-click on an existing column heading and select Column Info
or
• Double-click above the column name.
Figure 5.2 The Column Info Window
The Column Info window contains the following information:
Column Name
Type or edit the name of the column.
Lock
Lock the column so that none of its values can be edited. After you lock a column, the lock icon () appears next to the column name in the data table’s Columns panel. If you add a formula to a column, the column automatically locks.
Data Type
Select or change the data type of a column, which determines the following:
– How the column’s values are formatted in the data grid
– How the column’s values are saved internally
– Whether the column’s values can be used in calculations
Choose from the following data types:
– Numeric columns contain only numbers (with or without a decimal point).
– Character columns contain any characters, including numbers. In character columns, numbers are seen as characters and are treated as discrete values instead of continuous values. The maximum field width for character values is 32,766 bytes.
– Row State columns contain row state information, which indicates whether the rows are excluded, hidden, labeled, colored, or marked. See Row State Columns.
Note: Short-integer formats might also be available. See The Short-Integer Format.
Modeling Type
(Numeric or Character data types only) Select or change the modeling type of a column, which tells JMP how to treat the column’s values during analyses. You can change the modeling type to look at a variable in different ways. See About Data Types and Modeling Types.
Choose from the following modeling types:
– Continuous columns contain only numeric data types. Continuous values are treated as continuous measurement values. JMP uses the numeric values directly in computations.
– Ordinal columns contain either numeric or character data types. JMP analyses treat ordinal values as discrete categorical values that have an order. If the values are numbers, the order is the numeric magnitude. If the values are character, the order is the sorting sequence.
– Nominal columns contain either numeric or character data types. All values are treated in JMP analyses as if they are discrete values with no implicit order.
Format
(Numeric data types only) Select or change the display format of a numeric column. See Numeric Formats.
Initialize Data
(Appears only during new column creation) Specify the type of initial data values that you want to appear in the column. See Fill in Initial Data Values.
Column Properties
(Contains a list of different properties) Assign properties to columns. See Assign Column Properties.
Tip: Use the Next button to continue adding columns.
About Data Types and Modeling Types
A column in a JMP data table can contain different types of information. However, all information in a single column must have the same data and modeling types.
• When you import data, JMP guesses which data and modeling types to use. Therefore, you should verify that JMP has guessed correctly.
• When you manually insert data into JMP, you should assign a data type and a modeling type at that time.
Figure 5.3 illustrates the icons that identify the different modeling types.
Figure 5.3 Modeling Type Icons in the Columns Panel
Click on an icon to change the modeling type.
Tip: You can select Continuous only if your data type is numeric. If the Continuous option is dimmed on the menu and you want to make the column continuous, you must first change the column’s data type in the Column Info window.
The Short-Integer Format
When you use the correct short-integer format for your data, you do not see any difference in how the numbers appear, but the numbers occupy less disk space and use less memory. Short-integer formats must be activated in preferences to appear in the Column Info window.
To make short-integer formats available in the Column Info window:
1. Select File > Preferences and click Tables.
2. Select the Allow short numeric data format option.
3. Click OK to return to the data table.
To store numeric data in short-integer format:
1. Double-click above the column name whose values you want to be short-integer.
The Column Info window appears.
2. From the Data Type menu, select 1-byte integer, 2-byte integer, or 4-byte integer.
JMP stores values as integers in the range that you selected. The following numbers are examples:
• For 1-byte integer, the range of numbers that you can enter is from -126 to 127.
• For 2-byte integer, the range of numbers that you can enter is from -32,766 to 32,767.
• For 4-byte integer, the range of numbers that you can enter is from -2,147,483,646 to 2,147,483,647.
Numeric Formats
For numeric columns, the Format menu appears in the Column Info window. Specify the format to tell JMP how to display numbers in the column.
Note the following information:
• For all format options, you can specify the number of total characters that you want the cells in the column to accommodate. See Specify Width.
• For descriptions of the format options, see Numeric Format Options.
• To add commas to values that equal a thousand or more, select the Use thousands separator option. You must account a space for each comma in the Width box, or else they might not appear. This option is available for the Best, Fixed Dec, Percent, and Currency formats.
Specify Width
When you specify a number in the Width field, be sure to include the total number of possible characters. Characters include: numbers, decimal points, commas, and currency symbols.
Numeric Format Options
Choose from the following numeric format options:
Best
Allows JMP to consider the precision of each cell value and select the best way to show it. By default, the physical width of the column is 12 characters.
Fixed Dec
Shows all values in the column rounded to the number of decimal places that you specify.
Note the following:
– To see only whole numbers, set the number of decimal places to zero.
– If the value does not have as many numbers after the decimal as the number that you have specified, zeros are added to the number to give it that many decimal places. For example, if the value is 1.23 and you type 5 in the Dec box, JMP shows the number with five decimal places: 1.23000.
Percent
Multiplies numeric values by 100 and shows the number followed by a percent sign.
PValue
Shows probability values. The default value of the width is 12. If a number is less than 0.0001, the number is displayed as <.0001. The format is mostly used in JSL scripts and rarely needed for a data table column.
Scientific
Shows a number in standard scientific notation. If you enter the number 123456, it appears as 1.23456e+5.
Currency
Formats values with two decimal positions, thousands separators, and the currency sign that is specified in your computer’s locale settings (for example, on an American system, the number 123456 shows in the data table as $123,456.00). The default width of the Currency format is 15. If you have a number that requires a wider field width, the format defaults to the Best format. Once assigned, the currency symbol appears in the column and in graphs that contain the column.
Date
Shows all values in the column as a date. See Date Formats.
Time
Shows all values in the column as a specific instance in time, such as 12/2/03 at 2:23 PM. See Time Formats.
Duration
Shows all values in the column as a duration of time, such as hours, minutes, and seconds.
– :day:hr:m, :day:hr:m:s show a duration of time, such as 52:03:01:30, or fifty-two days, three hours, one minute, and thirty seconds.
– hr:m, hr:m:s, min:s shows a duration of time, such as 17:37, or seventeen hours and thirty-seven minutes.
Geographic
Shows latitude and longitude number formatting for geographic maps. Latitude and longitude options include the following:
– DDD (degrees)
– DMM (degrees and minutes)
– DMS (degrees, minutes, and seconds)
In each format, the last field can have a fraction part. You can specify the direction with either a signed degree field or a direction suffix. To show a signed degree field, such as -59°00'00", deselect Direction Indicator. To show the direction suffix, such as 59°00'00" S, select Direction Indicator.
To use spaces as field separators, deselect Field Punctuation. To use degrees, minutes, and seconds symbols, select Field Punctuation.
Date Formats
When you choose a Date format, you can also specify an Input Format. The Date format indicates how the date appears in the data table cells, and the Input Format indicates how you type in the date.
If you assign a date format to a numeric column that already contains data, then the numeric values are treated as the number of seconds since January 1, 1904. For example, if you have a numeric column with a cell value of 1,234,567,890 and you change the format to Date > m/d/y, the cell value appears as 02/13/1943.
The examples in Table 5.1 use the date of December 31, 2004.
 
Table 5.1 Date Formats 
Format
Appears As
m/d/y
12/31/2004
mmddyyyy
12312004
m/y
12/2004
yyyyQq
2004Q4
d/m/y
31/12/2004
ddmmyyyy
31122004
ddMonyyyy
31Dec2004
Monddyyyy
Dec312004
y/m/d
2004/12/31
yyyymmdd
20041231
yyyy-mm-dd
2004-12-31
Date Long
Friday, December 31, 2004
Date Abbrev
Dec 31, 2004
Locale Date
Varies based on local OS setting. Here is an example: in the United States, the local OS setting is mm/dd/yyyy (12/31/2004).
Note: To change the way a date appears in a graph without changing the way it appears in a data table, see Change the Numeric Format of an Axis in JMP Platforms.
Time Formats
When you choose a Time format, you can also specify an Input Format. The Time format indicates how the time appears in the data table cells, and the Input Format indicates how you type in the time.
• You can add the number of hours, minutes, and seconds after midnight of the prepended date for the following date formats:
– m/d/y
– d/m/y
– y/m/d
– ddMonyyyy
– Monddyyyy
– Locale Date
For example, December 31, 2004 has a numeric value of 3,187,296,600, which represents 12/31/2004 12:10 AM.
• :day:hr:m and :day:hr:m:s show the number of days, hours, minutes, and seconds since January 1, 1904. For example, the results for December 31, 2004 are :36890:00:10: and :36890:00:10:00.
• h:m:s and h:m show the hours, minutes, and seconds portion of the date in the date field. For example, the results for December 31, 2004 at 12:10 AM are 12:10:00 AM and 12:10 AM.
• yyyy-mm-ddThh:mm and yyyy-mm-ddThh:mm:ss show the year, month, day, and time (for example, 2004-12-31T12:10:00. T is a literal, representing itself).
Note: To change the way a time appears in a graph without changing the way it appears in a data table, see Change the Numeric Format of an Axis in JMP Platforms.
International Formats
If you are importing or entering data that contains formatting specific to country standards, you might need to make sure that your number formats are interpreted correctly. On Windows, access the Control Panel’s region and language option, and select the country for which the number should be formatted. On the Macintosh, from the Apple menu, select System Preferences > Language & Text > Formats, and select the correct country.
Row State Columns
Similar to assigning row states to rows, you can create a column that contains only row state information. A row state column stores information about whether rows are excluded, hidden, labeled, colored, marked, or selected. To designate a column as a row state column, in the Column Info window next to Data Type, select Row State.
Figure 5.4 Row States in Rows and a Row State Column
Since row state columns store the row states, you can apply them again later. Populate row state columns by copying them from the current row states or with column formulas.
To create a row state column:
First, create a row state column:
1. Select Cols > New Column.
2. Next to Data Type, select Row State.
3. Click OK.
Populate the cells with new row state information or copy existing row state information from rows.
To populate cells with new row state information:
1. To populate only certain rows in the row state column, highlight those rows. Or, to populate all rows in the column, highlight the row state column.
2. Right-click and select Row States Cells.
3. Select the row state that you want to apply.
To copy existing row state information:
1. To populate only certain rows in the row state column, highlight those rows. Or, to populate all rows in the column, highlight the row state column.
2. Click the star icon () beside the column name in the Columns panel.
3. Select one of the following:
– Copy from Row States Replaces the row states in the column with the row states from the rows.
– Add from Row States Adds the row states from the rows to the row state column.
– Copy to Row States Replaces the row states in the rows with the row states from the column.
– Add to Row States Adds the row states from the row state column to the row states in the rows.
Permanently Select Cells
You can save a selection in a row state column just like you save other row state characteristics (hide, exclude, color, and so on). This places a “permanent” highlight on a cell.
To permanently select cells:
1. Right-click a cell and select Row States Cells > Select/Deselect.
2. Repeat this for as many cells as you would like to select.
3. To remove the highlight, right-click on the cell and select Row States Cells > Select/Deselect.
Fill in Initial Data Values
When you add a new column to a data table, the Initialize Data menu appears in the Column Info window. Specify the type of initial data values that you want to appear in the new column. Select one of the following options:
Missing/Empty
Places missing values in the column, represented by a black dot (•) for numeric data and a blank space for character data.
Constant
Places one number or character in all of the column’s rows. Type the number or character into the box that appears. Enter any number of characters.
Today
Places today’s timestamp in the column for each row. This option is relevant only for the Date or Time formats.
Sequence Data
Inserts sequential data based on the parameters that you specify. See Numeric or Character Sequence Data.
Random
Inserts random data into the column. Select the type of random number that you want to use, and then enter one of the following:
– A range for random integers or random uniform numbers.
– The mean and standard deviation for random normal numbers.
– Values and proportions for random indicators.
Numeric or Character Sequence Data
To insert sequential data for numeric data:
1. Next to Data Type, make sure Numeric is selected.
2. Next to Initialize Data, select Sequence Data.
3. In the From and To boxes, assign a starting and ending point.
4. In the Step box, assign the sequence.
5. (Optional) In the Repeat each value N times box, type the number of times that you want each numeric value repeated.
6. Click OK.
For example, if you want the column to contain even numbers from 2 to 60, type 2 in the From box, 60 in the To box, and 2 in the Step box.
To insert sequential data for character data:
1. Next to Data Type, make sure Character is selected.
2. Next to Initialize Data, select Sequence Data.
3. In the box next to Add, type the character data and click Add.
4. (Optional) In the Repeat each value N times box, type the number of times that you want each character value repeated.
5. Click OK.
Assign Column Properties
Columns can contain special column properties, such as formulas, notes, and restrictions on values.
When you add a property to a column, the properties icon appears next to the column name in the Columns panel. Note the following exceptions:
•  Indicates that the range or list check property is applied.
•  Indicates that the column contains a formula.
• Columns with the Notes property do not contain an icon.
To assign a property to a selected column:
1. Right-click on the column that you want to assign a property to.
2. Select Column Properties and select the property that you want to assign.
Note: You can also right-click on a column and select Column Info. Select the property that you want to assign from the Column Properties menu.
The following sections describe the properties that you can add to columns.
Formula
Insert a formula into a column to compute the values for that column. After a formula is added, the column is locked so that its data values cannot be manually edited (preventing invalidation of the formula).
• Click Edit Formula to create a formula. For details about creating a formula, see the Formula Editor.
• If you do not want JMP to evaluate the formula, click Suppress Eval.
• If you do not want JMP to alert you about errors in your formula, click Ignore Errors.
• Once you have created a formula:
– In the Column Info window, a visual of the formula appears at right. However, if your formula is long, only a portion of it might appear. Click and drag the borders of the formula box to resize it.
– From the data table, edit the formula by clicking () next to the column name in the Columns panel.
Tip: To bypass the Column Info window when creating a formula, right-click on the column and select Formula.
Notes
Adds notes to the selected column.
Range Check
Range checking validates the data in a column. Set up the column to accept only numbers that fall within a specified range.
Select which formula to use to set up the range. x is the value entered into the column, a is the beginning of the range, and b is the end of the range.
• a = the lowest value that the column accepts
• b = the highest value that the column accepts
• For a single-sided range check, leave either a or b empty.
• From the data table, modify the range check by clicking () next to the column name in the Columns panel.
To turn off range checking:
1. From the data table, right-click the column name in the Columns panel.
2. Select Validation > No Checking.
List Check
List checking validates the data in a column. Set up the column to accept only individual numbers that you specify. List checking is useful when you want to specify how to order the data in your graphs or plots.
• Use the buttons to add new values, change, or reverse the order of values, and remove values.
• Once a list check is set on a column, the cursor changes to when positioned over the cells. If you try to enter a value not included on the validation list, a warning message appears.
• To see a menu of acceptable values, right-click a cell and select List Check Values. You can select the cell value from the menu instead of typing it into the cell.
• From the data table, modify the list check by clicking next to the column name in the Columns panel.
To turn off list checking:
1. From the data table, right-click the column name in the Columns panel.
2. Select Validation > No Checking.
Missing Value Codes
Use missing value codes to specify column values that should be treated as missing. For example, sometimes the value 99 is used as a placeholder to represent missing values, or perhaps several values are used to represent different types of missing values.
Value Labels
Use value labels to show a label in the data table instead of a value. A label appears for each instance of the value. You can show the original values by double-clicking a label within a cell.
• Type the value that you want to assign a label to in the Value box.
• Type the label that you want to appear in the Label box.
• To use ranges, click Allow Ranges then specify the lower and upper values.
Tip: To assign a label to missing values, enter a period (.) for the lower bound and leave the upper bound empty. To assign a label to all other values, enter three periods (...) for the lower bound and leave the upper bound empty.
• Add, change, or remove labels.
Note the following tips:
• To turn off value labels in the data table without deleting the value labels that you have set up, deselect Use Value Labels.
• When your data table contains value labels, using the Search commands searches for actual values, but does not search for labels.
• When your data table contains value labels, the Row Editor displays the label, and when the cell is highlighted for editing, it shows the actual value.
• If you copy and paste a cell with a value label, the actual value is pasted.
• In a formula, when you reference a column using value labels, hover your mouse over the value label to see the actual data value.
Value Ordering
Value ordering reorders categorical values.
Note: If you use both the Value Ordering and Row Order Levels properties, the Value Ordering property overrides the Row Order Levels.
The following values automatically appear in the appropriate order in reports:
• January, February, March, April, May, June, July, August, September, October, November, December
• Jan, Feb, Mar, Apr, May, Jun, Jul, Aug, Sep, Oct, Nov, Dec
• Sunday, Monday, Tuesday, Wednesday, Thursday, Friday, Saturday
• Very Low, Low, Medium Low, Medium, Medium High, High, Very High
• Strongly Disagree, Disagree, Neutral, Indifferent, Agree, Strongly Agree
• Failing, Unacceptable, Very Poor, Poor, Bad, Acceptable, Average, Good, Better, Very Good, Excellent, Best
Value Colors
Use value colors to assign the values of a nominal or ordinal column a certain color or range of color themes. The column’s values appear with the assigned color in all applicable graphs, such as mosaic plots and plots with color-coded legends. You can also color the values in the data table column.
• To change the color of a specific value, right-click a color circle and select a color.
• To use a color theme, select it from the Color Theme menu.
• To create a custom color theme, see Create a Custom Color Theme.
• To also color the cells in the data table, select Color Cell by Value.
• (Optional) Select from the options in the Macros menu. See Macros Options.
Create a Custom Color Theme
To create a custom color theme:
1. Select Custom from the Color Theme menu.
2. Create the color using the sliders.
3. Name the color and click Save.
4. Click OK two times.
You can access the new custom color in the Color Theme menu the next time you use the Value Colors property. The color is also saved in your preferences. See Create Color Themes in Enter and Edit Data.
Macros Options
The Macros menu contains the following options:
Gradient between ends
Sets the colors of the top and bottom values. JMP applies a color gradient across the entire range of values. Use this command to make all of the colors in between for the other levels.
Gradient between selected points
Sets the colors of the top and bottom values so that JMP can apply a color gradient to a range of values that you have highlighted in the Value Colors list.
Reverse colors
Reverses the color of the values from top to bottom or bottom to top.
Revert to old colors
Sets the colors back to their original color values.
Color Gradient
Select a color gradient to color a continuous column in a plot. Color gradients are supported in the Graph Builder, Bubble Plot, Tree Map, and Cell Plot platforms.
• To also color the cells in the data table, select Color Cell by Value.
• Select a color gradient from the menu.
• Enter the minimum, maximum, and center values:
– Minimum values reflect the color at the left of the gradient.
– Maximum values reflect the color at the right of the gradient.
– Center values reflect the color in the middle of the gradient.
Note: To see color gradients in Graph Builder, you must assign the column to the Color zone. To see color gradients in Bubble Plot and Tree Map, you must assign the column to the Coloring role.
Axis
Use the Axis property to change the default axis settings for a column. JMP automatically uses your settings when the column appears in an analysis.
Specify the following properties in the Axis panel:
Scale Type
Change the scale type to Linear, Log, Geodesic, or Geodesic US.
Min
Set the minimum value in the graph.
Max
Set the maximum value in the graph.
Minor Ticks
Specify the number of minor tick marks in the graph.
Inc
Specify the number of increments in the graph.
Show Major Ticks
Show major tick marks in the graph.
Show Minor Ticks
Show minor tick marks in the graph.
Show Major Grid
Show major gridlines in the graph.
Show Minor Grid
Show minor gridlines in the graph.
Show Labels
Show labels in the graph.
Orientation menu
Change the orientation of the axis labels. Choose from the following options:
– Automatic adjusts the orientation based on the length of the label text.
– Horizontal and Vertical describe orientations for single axes.
– Perpendicular and Parallel describe orientations for paired axes (for example, in Scatterplot Matrices).
– Angled adjusts the orientation to be angled at a 45 degree angle.
To set default axis properties for a column from within a graph:
1. Create the graph.
2. Change the axis to your preferred specifications. See Customize Axes and Axis Labels in JMP Platforms.
3. Right-click the axis and select Save to Column Property.
Coding
When you fit a column in a model, the low and high values of the column are transformed to –1 and 1, which makes tests and parameter estimates more meaningful. This is called coding. Use coding to specify which values you want to use as the low and high values in a column. Coding can be used for any continuous variable. Coding is the default for continuous factors generated by the DOE commands in JMP.
Note: If a column has one or more limits missing, JMP substitutes the data’s minimum and maximum for the high and low values.
Mixture
You might have a column in a data table that is one of several factors that form 100% of a mixture. Use the Mixture property to set up the column so that JMP uses it to automatically generate a no-intercept model (using the Fit Model command) when you analyze the data.
• Enter the lower and upper limits and the sum of terms.
• (Optional) Select the L and U PseudoComponent Coding options.
Note: For more information, see the Custom Design chapter in Design of Experiments.
Row Order Levels
The row data in an analysis report appears in order of the data’s values. Use the Row Order Levels property to set the row data to appear in the same order in which it appears in the data table column.
The row ordering applies only to the selected column. To apply it to other columns, repeat the above steps for each column, or use the Standardize Attributes command. See Standardize Attributes and Properties across Columns.
Tip: To show the analyzed row data in another order (besides according to their values or their occurrence in the data table columns) use the Value Ordering property. See Value Ordering. The Value Ordering property overrides the Row Order Levels property when both are evoked.
Spec, Control, and Response Limits
Use the Spec Limits, Control Limits, and Response Limits properties to save specification, control, and response limits in a column. When you perform a capability analysis, control chart analysis, or analyses displaying the prediction and contour profilers, you do not have to re-specify roles and limits each time. Saving these limits in a column also facilitates consistency from use to use. For example, you might run an analysis that uses these limits. When you come back later and change the data, you can run a new analysis on the new data using the same limits.
Specification Limits
Specification limits are used when you perform a capability analysis using the Distribution and Capability platforms.
• Enter a lower and upper specification limit and a target value for a numeric column.
• The Show as graph reference lines option draws specification limits as reference lines on a graph.
Control Limits
Control limits are used when you create a control chart.
• Select the control chart type.
• Enter the values for the average and lower and upper control limits. If any of these values are missing, JMP includes a calculated value in the control chart.
Response Limits
Response limits are used by the prediction and contour profilers and DOE.
• Specify one of the possible goals for a DOE response variable. Choose from the following: Maximize, Match Target, Minimize, or None. For example, if you are in the Prediction Profiler, and you want the desired value to be close to zero, select Match Target.
• If you have two responses, you can enter an Importance value, which specifies how to weigh the importance of one response against the other response.
• Specify values for the lower, middle, and upper limits and desirability values.
• To draw specification limits as reference lines on a graph, select the Show as graph reference lines option.
Design Role
Specify a design roll to indicate how to use a factor column in a model to design an experiment.
Factor Changes
To create a split plot design using the JMP DOE commands, you must identify a factor as hard, meaning that the factor would be difficult to change. You can identify a factor as hard in the DOE design pane each time you design an experiment. To create a split plot design every time you use a certain factor, you can save time by setting up that factor to be hard in all experiments.
Sigma
Use the Sigma property to enter a known sigma value. This value is used by applications such as control charts or any application that requires a sigma value to complete computations. If no sigma value is supplied, sigma is calculated from the sample.
Units
Use the Units property to specify the measurement units that were used to collect the data for the column. The units appear in parenthesis after the column name. For example, you might want a column to indicate that age values are measured in months, or that a monetary value is in thousands of dollars.
Distribution
For a column that contains continuous numeric data, use the Distribution property to select a distribution type to fit to the column. When you run the Distribution report (Analyze > Distribution) for the column, JMP automatically estimates a fit using the specified distribution. A curve reflects when the data completely fits the specified distribution.
Set the Distribution property only when you already know how the data is distributed. For example, you might already know before you run Analyze > Distribution that the data has a Weibull distribution.
If you set the Distribution property and the Spec Limits property, then the Distribution report produces a Capability Analysis report, reflecting the distribution type that you selected for the column.
Time Frequency
When using the Time Series platform, you can assign the Time Frequency property to data. The Time Frequency property specifies the frequency with which the data is reported (such as annually, quarterly, monthly, and so on). Specifying a time frequency allows JMP to take things like leap years and leap days into account. If no frequency is specified, the data is treated as equally spaced numeric data.
Map Role
If you have created your own data table that contains boundary data (such as countries, states, provinces, or counties) and you want to see a corresponding map in Graph Builder, use the Map Role property.
Note the following:
• If the custom boundary files reside in the default custom maps directory, then you need to specify only the Map Role property in the -Name file.
• If the custom boundary files reside in an alternate location, then you must specify the Map Role property in the -Name file and in the data table that you are analyzing.
• The columns that contain the Map Role property must contain the same boundary names, but the column names can be different.
Note: For an example using the Map Role property, see the Graph Builder chapter in Basic Analysis and Graphing.
To add the Map Role property into the -Name data table:
1. Right-click on the column containing the boundaries and select Column Properties > Map Role.
2. Select Shape Name Definition.
3. Click OK.
4. Save the data table.
To add the Map Role property into the data table that you are analyzing:
Note: Perform these steps only if your custom boundary files do not reside in the default custom maps directory.
1. Right-click on the column containing the boundaries and select Column Properties > Map Role.
2. Select Shape Name Use.
3. Next to Map name data table, enter the relative, or absolute path to the -Name map data table.
If the map data table is in the same folder, enter only the filename. Quotes are not required when the path contains spaces.
4. Next to Shape definition column, enter the name of the column in the map data table whose values match those in the selected column.
5. Click OK.
6. Save the data table.
When you generate a graph in Graph Builder and assign the modified column to the Shape zone, your boundaries appear on the graph.
Create Your Own Column Property
You can create your own column property and assign it any name that you choose. This property is then available for JSL programming.
1. Right-click on the column and select Column Properties > Other.
2. Type a name for the new property.
3. Enter a value for the property.
Response Probability
The Response Probability property does not appear on the Column Properties menu. JMP automatically assigns this property when the following steps are performed:
1. Create a logistic regression using either the Fit Y by X platform or the Fit Model platform.
2. Select the Save Probability Formula option in the logistic report window.
JMP automatically assigns the Response Probability property to the new probability columns. The Response Probability property makes all of the levels of the categorical variable appear as a single row (instead of separate rows) in the Profiler report.
Note: For more details, see the chapter on profiling in the Modeling and Multivariate Methods book.
Standardize Attributes and Properties across Columns
A column might contain attributes (data types, modeling types, numeric formats, and so on) or properties (formulas, notes, list and range checks, and so on) that you want other columns to have. You can use the existing column to standardize the attributes and properties across columns. This includes both adding and deleting attributes and properties.
Add Attributes and Properties
To apply an existing column’s attributes and properties to multiple columns:
1. Select the column containing the desired attributes or properties.
2. Select Cols > Standardize Attributes. The window in Figure 5.5 appears.
Figure 5.5 Standardizing Attributes Across Columns
By default, the items within the Standardize Attributes panel are dimmed. To access an item:
3. Click the Attributes button and select the items to be duplicated across columns.
Note: The Input Format item is applicable only for the Date, Time, and Duration formats.
4. To change the values of any of the attributes, use the menus in the Standardize Attributes panel.
5. Click Column Properties in the Standardize Properties area and set up the properties that you want the other columns to have.
6. Click in the data table.
7. Highlight the columns in the data table to which you want to apply the attributes and properties.
8. Go back into the Standardize Columns Attributes window and click Apply. The attributes and properties apply to all the selected columns in the data table. You can view your results.
9. Click OK.
Tip: You can select all of the columns that you want to standardize and then select Cols > Standardize Attributes. The first selected column (left-most) is used to set up the window’s values. Then you might skip steps 6 and 7, because you have already selected the columns.
Delete Properties
To delete the same properties across multiple columns:
1. Select the column containing the attributes or properties that you want to delete.
2. Select Cols > Standardize Attributes.
3. Click Column Properties in the Delete Properties area and select the properties that you want to delete.
4. Click OK.
Example of Standardizing a Formula
If you have applied a formula to a column, and you want to apply that same formula to additional columns in the data table, use the Substitute Column Reference option.
Note: This option is dependent upon the location of the column that is referenced in the original formula. For example, if your original formula is based on the previous column, then any other formulas applied to additional columns are based on their previous columns.
For example, the Blood Pressure.jmp sample data table contains blood pressure measurements taken on five subjects three times each day, over a period of three days. You want to find the log of each blood pressure (BP) column.
1. Open the Blood Pressure.jmp sample data table.
Create nine new columns, one for each existing BP column.
2. Select Cols > Add Multiple Columns.
3. Add nine columns.
4. Click OK.
Apply your original formula as follows:
5. Right-click on Column 1 and select Formula.
6. Select BP 8M.
7. Select Transcendental > Log.
8. Click OK.
Column 1 now contains the log of the BP 8M column. You want the rest of the empty columns to contain the log of the remainder of the BP columns.
9. In the data table, select all of the new columns that you created, including the one with the original formula (columns 1-9).
10. Select Cols > Standardize Attributes.
11. In the Standardize Properties panel, click on Column Properties and select Formula.
12. Select the check box next to Substitute Column Reference.
13. Click OK.
Now all of the new columns are populated with the log of the BP columns, in the order in which they appear. Column 1 contains the log for BP 8M, Column 2 contains the log for BP 12M, and so on.
Assign a Preselected Analysis Role
You can assign an analysis role, such as x, y, weight, or frequency, to a selected column and save the role with the data table. When you do this and then run an analysis, JMP uses the preselected role to automatically fill in the role boxes in windows. Then you do not have to specify these roles each time you run an analysis. For example, you might want a column named height to take the x role in every analysis of that data table. To enforce the x role, you assign the preselected role of x to the column.
When you select Freq, the values in that column are what JMP uses as the frequency of the observation. If n is the value of the Freq variable for a given row, then that row is used in computations n times. If it is less than 1 or is missing, then JMP does not use it to calculate any analyses.
When you select Weight, the values in that column provide weights for each observation in the data table. The variable does not have to be an integer, but it is included only in analyses when its value is greater than zero.
To assign a preselected role to a column:
1. Highlight the column.
2. Select Cols > Preselect Role.
3. Select a role: No Role, X, Y, Weight, or Freq.
After you select the appropriate roles, icons in the Columns panel signify what roles have been assigned. Click the icon to access a list of roles and select a different one. See Icons Representing Column Characteristics and Properties in Get Started.
Compress Selected Columns
JMP lets you compress columns in a data table to minimize the size of the file and reduce the amount of memory required to analyze data. This feature is helpful when numeric columns contain many small integers or when any column contains fewer than 255 unique values. For example, compressing columns in a data table with 389 columns and 85,000 rows might decrease the file size from 250MB to 33MB, depending on the type of data.
When you compress columns, JMP verifies whether the data can be stored in a more compact form based on the data type:
• In character columns with fewer than 255 unique values, the List Check property is added to the column (shown in Figure 5.6).
• In numeric columns, data is compressed to 1-byte, 2-byte, or 4-byte integers when possible (shown in Figure 5.7). For details about short integers, see The Short-Integer Format.
A numeric column with non-integer values can also be compressed if there are fewer than 255 unique values. In this case, the List Check property is added to the column.
Caution: In a column with the List Check property, you can enter only a value that is in the list. Otherwise, JMP warns that the cell contains invalid data when you try to enter the new value. For details about list checking, see List Check.
Figure 5.6 List Check Property Added to a Compressed Character Column
Figure 5.7 Column Info Window Showing Numeric Column Before and After Compression
To compress columns, select one or more columns and select Cols > Compress Selected Columns. (Select all columns if you do not know which columns can be compressed.)
The column or columns are compressed if possible. The log shows which columns were compressed and how they were compressed. (Select View > Log to show the log.)
Note: To compress a numeric column manually, set your Tables preferences to allow short numeric data and then change the column’s data type to 1-byte integer, 2-byte integer, or 4-byte integer. For details about this preference, see Tables in JMP Preferences.
 
..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
3.149.241.98