Transformations of Variables - Review and Plot Tab
Select the Review & plot tab of the Transformations of Variables dialog box to access options to review or plot the currently highlighted variable (together with other variables). The labeling of the individual observations (that is, of the respective points in time represented by the rows in spreadsheets or the horizontal x-axis in plots) is determined by the selections in the Label data points with group box (see below). For the plots, you can choose to Scale the x-axis in the plots manually (see below). This is often useful when the series consists of several regularly recurring time intervals, for example, if it is made up of 10 years with 12 (monthly) observations each. Then it may be useful to scale the x-axis in the plot so that each major tickmark indicates one year.
Element Name | Description |
---|---|
Review highlighted variable/Plot | Click the Review highlighted variable button or adjacent Plot button to display a spreadsheet or a plot, respectively, containing the values for the currently highlighted variable. |
Review multiple variables/Plot | Click the Review multiple variables button or adjacent Plot button to display a standard variable selection dialog, in which you select a list of variables to be displayed or plotted. When plotting multiple variables the vertical (Y) axis will have a common scale; thus, if the series that are plotted contain values of completely different magnitudes use option Plot two var lists with different scales, instead. |
Plot two var lists with different scales | Click the Plot two var lists with different scales button to plot two lists of variables using different scales for the vertical (Y) axes in the plot. The first list is plotted against the left y-axis and the second list is plotted against the right y-axis. For example, suppose you want to plot two variables (series) called S1 and S2, and that these variables contain values of different magnitudes: the values of S1 range from 1-100 and the values of S2 range from 1000 to 10000. If you plot S1 against the left y-axis (scaled from 1-100) and S2 against the right y-axis (scaled from 1000-10000) then both variables will be plotted with optimal resolution, and thus relationships between patterns across time (the x-axis) may become more clearly identifiable. |
Plot variable (series) after each transformation | Select the Plot variable (series) after each transformation check box to plot the transformed variable after each transformation. |
Display/plot subset of cases only | If the Display/plot subset of cases only check box is selected, the from and through boxes become available, in which you enter the values to select a range of observations to include in the spreadsheet or plot. |
Scale X axis in plots manually (min, step) | The setting of this check box affects the scaling of the x axis in line plots of single or multiple variables (series). The choice of scaling will be retained across other analytical procedures in this module, for example, when plotting forecasts from an ARIMA analysis. By default, if this check box is not set, the x-axis in all plots will always be scaled so that (1) all valid data points are displayed in the plot, and (2) that the tickmarks (steps) on the x-axis are "neat" (i.e., the significant digit of the step size is either 1, 2, or 5). However, if there is a known periodicity in the data that is of interest, such as 12 months per year or 10 observations per day, etc., then it may be useful to select this check box and to scale the x-axis for plots accordingly. |
Label data points with | The Label data points with group box contains four options: Case names, Dates from a var, Case numbers, and Consecutive integers. The selected option button determines the labeling of the data points in spreadsheets or plots. The setting of these option buttons will be retained across other analytical procedures in this module, for example, when displaying residuals from an ARIMA analysis. |
Case names | Select the Case names option button if there are case names available in the data file to use the names to label the data points (rows in spreadsheets and the x-axis in plots). |
Case numbers | By default, the Case numbers option button is selected. In this case, each point will be identified by the respective case number, and case numbers will be used on the x-axis in plots of available series. |
Dates from a var | Select the Dates from a var option button to display a standard variable selection dialog, in which you select the variable containing dates (i.e., values formatted as dates). Note you can also open this dialog by clicking the Variable button (see below). Those dates will then be used to label points in spreadsheets and the x-axis in plots. |
Variable | Click the Variable button to display the standard variable selection dialog, in which you select the variable containing dates (i.e., values formatted as dates). |
Consecutive integers | Select the Consecutive integers option button to label, starting from 1, the points and x-axis in plots. Note that what will be displayed is the offset (plus one) of the respective data point from the first data point of any variable in the active work area. For example, suppose you have in the active work area two series called S1 and S2. S1 has its first valid observation at case number 5, and S2 has its first valid observation at case number 8. If you select the Review multiple variables button (see above) to display both variables, and select the Consecutive integers option button, then the first valid observation for variable S1 will be labeled 1, and the first valid observation for variable S2 will be labeled 4. |