|
|
|
|
Data Entry |
You may enter data directly into the STATGRAPHICS Mobile datasheet, copy it from Excel Mobile, or read it from a file created on your notebook or desktop computer. Microsoft ActiveSync will automatically keep data files on your mobile device and other computer in sync. |
|
|
|
|
|
|
|
Column Properties |
Every column has a name, comment, type and use. The Use property tells the program how the data should be treated when a statistical analysis is performed. |
|
|
|
|
|
|
|
Selecting a Statistical Procedure |
To run a statistical procedure, highlight the data you wish to analyze. Then select an entry from the Stats menu. The program will automatically select a statistical analysis based on the type of data you have highlighted. |
|
|
|
|
|
|
|
Output Pages |
Each output page consists of a graph plus selected numerical statistics. The output is formatted to display on the mobile device without using scrollbars. |
|
|
|
|
|
|
|
Additional Pages |
Each statistical procedure has additional pages of output that may be selected from the Output menu. |
|
|
|
|
|
|
|
Drill-Down Point Identification |
If you select a point with the stylus, information about that point will be displayed. This may include the point's coordinates, the number of standard deviations from the point to the fitted model, or other statistics. |
|
|
|
|
|
|
|
Session Log Files |
All output may be saved in log files. Back on your desktop, the STATGRAPHICS Mobile Assistant will convert the log files to RTF format, so that output may be copied to Word or PowerPoint. |
|
|
|
|
|
|
|
Capability Analysis |
Process capability can be estimated from a sample of measurements. Both short-term and long-term capability indices can be calculated, including the Sigma Quality Level. |
|
|
|
|
|
|
|
Data Transformations |
One of the column properties is the assumed distribution. If you select lognormal or power normal, then the data will be transformed appropriately before statistical techniques are applied that assume normality. |
|
|
|
|
|
|
|
Control Charts |
Statistical process control charts may be created for both variable and attribute data. Control limits may be determined from the data (Phase I) or specified by the user (Phase II). |
|
|
|
|
|
|
|
Entering New Data |
You may add additional data to the bottom of the datasheet and then select Recalc from the main menu to recalculate the latest analysis. |
|
|
|
|
|
|
|
Zooming and Translation |
Optional buttons allow any graph to be enlarged or shifted in any direction to obtain a better view of the most important data. |
|
|
|
|
|
|
|
Graphics Options |
You may change almost any feature on a graph, including point and line types, colors, scaling, and axis titles. |
|
|
|
|
|
|
|
Six Sigma Calculator |
The Tools menu includes a statistical calculator that can determine critical values for common probability distributions and a Six Sigma calculator than will translate amongst important quality performance measures. |
|
|
|
|
|
|
|
Variable Data |
A wide range of tests are available for variable data, including tests for normality. The program calculates confidence intervals, hypothesis tests, percentiles, and other summary statistics. |
|
|
|
|
|
|
|
Outliers |
Grubbs' test may be used to help determine whether a suspect data value is really an outlier. If so, it may be replaced by a missing value and the analysis recalculated. |
|
|
|
|
|
|
|
Time Series Forecasting |
Statgraphics Mobile estimates and forecasts time series data using seasonal and nonseasonal ARIMA models. It is ideal for updating your forecasts while away from your office. |
|
|
|
|
|
|
|
Two Sample Comparisons |
Data from two samples, either paired or unpaired, may be compared to determine whether or not they show a statistically significant difference. |
|
|
|
|
|
|
|
Multiple Comparisons |
Both Fisher's LSD and Tukey's HSD procedures may be used to compare data from three or more samples. |
|
|
|
|
|
|
|
Calibration and Stability Studies |
The curve-fitting procedure will determine prediction limits for Y given X or X given Y. One-sided bounds are useful for calibration and stability studies. |
|
|
|
|
|
|
|
Attribute Data |
Procedures for analyzing attribute data include a crosstabulation, various contingency table statistics, barcharts, mosaic plots, and a Pareto chart. |
|
|
|
|
|
|