Revised: 8/23/2012

Applies to:

- Minitab 16

- Quality Companion 3.3

Solution

The Move License feature allows a single-user license to be moved to a new computer without having to contact Tech Support. It deactivates the license on the current computer and generates a new Product Key to use on a new computer. 

Not all product licenses can be moved. If you meet one of the following criteria, you cannot use the Move License feature:

- You purchased an academic (OnTheHub.com) License

- You purchased a Timed or Fixed Duration License

- You activated the software using the Web Activation method

- Also, in Quality Companion the Move License feature is only available in versions 3.2.1 and later.

To Move a License:

- Launch Minitab or Quality Companion.

- Choose Help > Licensing to launch the Single-User License Utility.

- Click Move.

- Click Yes to confirm you want to deactivate the product on the current computer.

- Click Save to text file to save the new Product Key to a location on your computer. 

- Activate your software on a new computer using the Product Key created during the Move process. 

Note: If you purchased an upgrade version, the new key produced by Move License will also be an upgrade key. Therefore, you must install and activate a previous version of Minitab on the new computer prior to using the new key.

If multiple Minitab products are installed that are equipped with the Move License feature both products will be deactivated as part of one Move process. The new Product Key will activate both products on the new computer. For example, if you have movable licenses of both Minitab 16 and Quality Companion installed on the same computer, using the Move License feature in either product will deactivate both products, and produce a single new product key that will activate both products on a new computer.

In Minitab 16, you can apply a Box-Cox transformation to response values in a regression analysis using Stat > Regression > General Regression.

To apply a Box-Cox transformation in DOE or to create charts of the optimization process for the Box-Cox lambda, use the Box-Cox Transformation macro available from our Macros Library. This macro displays plots of: 

- The log-likelihood function with an estimate and 95% confidence interval for lambda
- The values of the PRESS statistic transformed back to the original units over the 95% confidence interval
- The influence of individual cases on the selection of lambda

1. Click the link, Box-Cox Transformation for Regression and Response Surface Models, below.

2. To the right of the Box-Cox Transformation for Regression and Response Surface Models description, click Code. (You can also click Documentation for more information about this test.)

3. Save the page as a macro, called Bctrans.MAC, in your Minitab macros directory (in Minitab 16, the default location is C:\Program Files\Minitab\Minitab 16\English\Macros). For instructions, see "How to save this page as a macro" at the top of the Code page.

For example, suppose the response variable is in C1, the predictors are in C2 and C3, and you want to display an index plot in addition to the default output.

In the Session window:
i. At the command prompt (MTB >) type %Bctrans C1 C2 C3; 
ii. Press Enter.
Iii. At the subcommand prompt (SUBC>), type influence.
Iv. Press Enter.

Note: To display the command prompt, activate the Session window and choose Editor > Enable Commands.

IMPORTANT: Do not use Stat > Control Charts > Box-Cox Transformation to transform the response variable in a regression or DOE setting because this procedure does not take into account the relationship between the response and the predictors.

Note: After applying a Box-Cox power transformation and analyzing the data in transformed units, you can create a contour plot to visualize the fitted regression or DOE model in the original units. For more information, see Knowledgebase ID 1959.

You can use Calc > Calculator to transform binomial data (proportions) or Poisson data (counts).
 

Binomial data (proportions)
You must have one column (or stored constant) for the number of trials (n) and one for the number of events (x). Trials must be positive integers, and events must be integers between 0 and n inclusive. 

For example, suppose your number of trials are in C1, your number of events are in C2, and you want the transformations stored in C3.

1. Choose Calc > Calculator.

2. In Store result in variable, enter C3.

3. In Expression, enter FTP(C1,C2).

4. Click OK.

The FTP command uses the transformation:
(arcsin(sqrt(np/(n+1))) + arcsin(sqrt((np+1)/(n+1))))/2

Poisson data (counts)
You must have one column (or stored constant) for the counts, which must contain nonnegative integers. 

For example, suppose the counts are in C4, and you want the transformations stored in C5.

1. Choose Calc > Calculator.

2. In Store result in variable, enter C5.

3. In Expression, enter FTC(C4).

4. Click OK.

The FTC command uses the transformation: 
(sqrt(x)+sqrt(x+1))/2

References:
Bisgaard, Soren and Howard T. Fuller (1994) Quality Quandaries, Analysis of Factorial Experiments with Defects or Defectives as the Response, Quality Engineering, 7(2), 429-443.

Freeman, M.F. and J.W. Tukey (1950) Transformations Related to the Angular and the Square Root, Annals of Mathematical Statistics, 21(4), 607-611.

Suppose that the variables you want to correlate are in columns C1 and C2:

1. Choose Stat > Tables > Cross Tabulation and Chi-Square.

2. In For rows, enter C1. In For columns, enter C2.

3. Click Other Stats.

4. Check Correlation coefficients for ordinal categories.

5. Click OK in each dialog box.

Minitab displays a value for Spearman’s rank correlation coefficient (Spearman’s rho) in the Session window.

You can also use Correlation to obtain Spearman's rank correlation coefficient. Suppose that the variables you want to correlate are in columns C1 and C2, and columns C3 and C4 are empty:

1. Delete any rows that contain missing values.

2. Choose Data > Rank.

3. In Rank data in, enter C1. In Store ranks in, enter C3. Click OK.

4. Choose Data > Rank.

5. In Rank data in, enter C2. In Store ranks in, enter C4. Click OK.

6. Choose Stat > Basic Statistics > Correlation.

7. In Variables, enter C3 C4. Click OK.

Minitab displays the Pearson correlation value for the ranked data, which equals the Spearman’s rank correlation coefficient for the original (unranked) data.

Reference: Gottfried E. Noether (1991). Introduction to Statistics-The Nonparametric Way. Springer-Velag, Inc.

Predicted R-squared is used in regression analysis to indicate how well the model predicts responses for new observations, whereas R-squared indicates how well the model fits your data. Predicted R-squared can prevent overfitting the model and can be more useful than adjusted R-squared for comparing models because it is calculated using observations not included in model estimation. Overfitting refers to models that appear to explain the relationship between the predictor and response variables for the data set used for model calculation but fail to provide valid predictions for new observations.
 

Predicted R-squared is calculated by systematically removing each observation from the data set, estimating the regression equation, and determining how well the model predicts the removed observation. Predicted R-squared ranges between 0 and 100% and is calculated from the PRESS statistic. Larger values of predicted R-squared suggest models of greater predictive ability.
 

For example, you work for a financial consulting firm and are developing a model to predict future market conditions. The model you settle on looks promising because it has an R-squared of 87%. However, when you calculate the predicted R-squared you see that it drops to 52%. This may indicate an overfitted model and suggests that your model will not predict new observations nearly as well as it fits your existing data.
 

Note: You can find this information in the Minitab Glossary. Choose Help > Glossary. On the Index tab, inType in the keyword to find, type r-sq, and then double-click on R-squared predicted in the list below.
 

To view the formula for the predicted R-squared, see Knowledgebase ID 922.
 

Note: You can get the predicted R-squared using Stat > Regression > Regression > Options or Stat > Regression > Stepwise > Options.

The Assistant in Minitab 16 was specially designed to help investigators in the field of quality to quickly select the most appropriate analysis for their data and to readily interpret the results. For this reason, we developed and implemented statistical methods that allow quality investigators to efficiently analyze their data, verify the assumptions for the analysis, and check the reliability of the results. The results are then presented in the Assistant’s uniquely designed Reports and Report Cards.
 

The methods used by the Assistant are based on established statistical practice and theory, referenced guidelines in the literature, and extensive simulation studies performed by statisticians in Minitab’s Research and Development department.

This error occurs when Minitab or Quality Companion is unable to read the license information.
 

To resolve this issue:
 

1. Click OK to close the "Cannot Retrieve License Information" error message.
 

2. Open a command prompt, using a method below.
 

In Windows XP:
      a. Click the Start button and choose Run.
      b. Type cmd and click OK.
 

In Windows Vista and Windows 7:
      a. Click the Start button and then click the Search Programs and Files box.
      b. Type cmd.
      c. Locate the command prompt in your search results, right-click on it, and choose Run as Administrator.
 

3. At the command prompt, type C: and press [Enter].
 

4. Do one of the following. If you have both Minitab 16 and Quality
Companion 3 you should only need to perform this step for one product.
 

For Minitab 16: 
At the command prompt, type:
      "C:\Program Files\Minitab\Minitab 16\Auth\haspdinst" -i and then press [Enter].

Note: If you are running a 64-bit version of Windows, type:
      "C:\Program Files (x86)\Minitab\Minitab 16\Auth\haspdinst" -i 

For Quality Companion 3: 
At the command prompt, type:
      "C:\Program Files\Minitab\Quality Companion 3\Companion\Auth\haspdinst" -i and then press [Enter].

Note: If you are running a 64-bit version of Windows, type:
      "C:\Program Files (x86)\Minitab\Quality Companion 3\Companion\Auth\haspdinst" -i
 

5. A dialog titled Sentinel HASP Run-time Environment Installer should appear. After the message in the dialog changes to Operation successfully completed, click OK.
 

6. Start Minitab 16 or Quality Companion 3.
 

If performing the steps above does not resolve the issue, please call us using the Need Assistance? number listed at the left of this page.