Migration Guide from ATP

In this chapter, we discuss differences between the analysis procedures and the Analysis Toolpack (ATP) ones. Please use the left sidebar to get more details on each command. Feel free to contact us if you have any questions.

ATP and StatPlus Features Matrix

Analysis Toolpak Command (click to see details)	StatPlus Command
ANOVA - Single Factor	Analysis of Variance (ANOVA) → One-way ANOVA
ANOVA - Two-Factor With Replication	Analysis of Variance (ANOVA) → Two-way ANOVA
ANOVA - Two-Factor Without Replication	Analysis of Variance (ANOVA) → Two-way ANOVA
Correlation	Basic Statistics → Linear Correlation (Pearson)
Covariance	Basic Statistics → Covariance
Descriptive Statistics	Basic Statistics → Descriptive Statistics
Exponential Smoothing	Time Series → Exponential Smoothing
F-Test Two-Sample For Variances	Basic Statistics → F-Test For Variances
Fourier Analysis	Time Series → Fast Fourier Transform - Direct
Histogram	Basic Statistics → Histogram
Moving Average	Time Series → Moving Average
Random Number Generation	[Data] → Random Numbers Generation
Rank and Percentile	Nonparametric Statistics → Rank and Percentile
Regression	Regression → Linear Regression
Sampling	[Data] → Random Sample [Data] → Conditional Sample [Data] → Periodic Sample
t-Test: Paired Two Sample for Means	Basic Statistics → Compare Means (T-Test)
t-Test: Two-Sample Assuming Equal Variances	Basic Statistics → Compare Means (T-Test)
t-Test: Two-Sample Assuming Unequal Variances	Basic Statistics → Compare Means (T-Test)
z-Test: Two Sample for Means	Basic Statistics → Two-Sample z-Test for Means

Details

ANOVA - Single Factor

Run the Analysis of Variance (ANOVA) → One-way ANOVA command from the Statistics menu.

ANOVA - Two-Factor With Replication

Like many statistical packages StatPlus uses the following data layout for the two-way ANOVA: two factor variables  and one response variable. If your data is arranged in a table - please use the Stack columns command to prepare data for the analysis.

Run the Analysis of Variance (ANOVA) → Two-way ANOVA command from the Statistics menu.

ANOVA - Two-Factor Without Replication

Run the Analysis of Variance (ANOVA) → Two-way ANOVA command from the Statistics menu.
Under the Options check the No Interaction(s) option.

Correlation

Run the Basic Statistics → Linear Correlation (Pearson) command from the Statistics menu.

Matrix with correlation coefficients, critical values and p-values for each pair of variables is produced.

Covariance

Run the Basic Statistics → Covariance command from the Statistics menu.

Descriptive Statistics

Run the Basic Statistics → Descriptive Statistics command from the Statistics menu.

Exponential Smoothing

Run the Time Series → Exponential Smoothing command from the Statistics menu.

To change the Damping factor please use the Alpha option.

Then select a variable with time series and click OK.

F-Test Two-Sample For Variances

Run the Basic Statistics → F-Test For Variances command from the Statistics menu.  One-sided p-levels for H1: F > 1 (Analysis Toolpak) and H1: F < 1, and two-sided p-level for H1: F ≠ 1 are calculated.

Fourier Analysis

Run the Time Series → Fast Fourier Transform - Direct command from the Statistics menu. Select variables with real (Re) and imaginary (Im) parts of the time series and click the OK button. Variable with an imaginary part is optional. If the variable with an imaginary part is not specified it is considered to be identically zero.

If you need to do inverse transformation - run the Time Series → Fast Fourier Transform - Inverse command from the Statistics menu.

Histogram

Run the Basic Statistics → Histogram command from the Statistics menu.
A histogram is plotted for each input variable. To get a Pareto histogram please use the Pareto histogram option in the Options. You can also select between right-closed intervals (Excel) or left-closed (right-open) intervals.

Moving Average

Run the Time Series → Moving Average command from the Statistics menu. Select a variable with time series and click the OK button.

Random Number Generation

Open the Random Numbers Generation menu from the Data menu. Select the distribution from the popup menu. Parameters for each procedure are similar to those in the Analysis Toolpak (ATP).

Rank and Percentile

Run the Nonparametric Statistics → Rank and Percentile command from the Statistics menu.

Regression

Run the Regression → Linear Regression command from the Statistics menu.
To add the residuals plot to the report please use the Plot Residuals option in the Options. You can also choose to get a collinearity diagnostics report, run heteroscedasticity tests and normality tests for residuals. Available residuals for plots are regular residuals (raw), standardized residuals, studentized residuals and deleted residuals (externally studentized residuals).

Sampling

All of the sampling methods are located under the Data menu. Run a command ( Random Sample, Conditional Sample or Periodic Sample) from the Data menu, select variables for sampling and click the OK button. Parameters for each procedure are similar to those in the Analysis Toolpak (ATP).
To change the condition for the Conditional Sample command, please use the Condition option.

 

t-Test: Paired Two-Sample for Means, Two-Sample Assuming Equal Variances, Two-Sample Assuming Unequal Variances

Run the Basic Statistics → Compare Means (T-Test) command from the Statistics menu.

To change the t-test type use the T-Test Type combo box in the Options.
To compare two samples using summarized data use the Compare Means (use summarized data) command.

z-Test: Two-Sample for Means

Run the Basic Statistics → Two-Sample z-Test for Means command from the Statistics menu.

One-sample z-Test is also available.