Descriptive Statistics

Central tendency : Mean, Median, Mode, Harm & Geom Mean
Variability or Dispersion : Range , Variance, Standard Deviation, Quartile, IQR, CI 
Distribution : Skewness, Kurtosis 
Normality Test : Shapiro Wilks, Kolmogorov- Smirnov & Liliefors, Anderson Daring, D'Agostino Pearson 
Chart : Histogram & Normal P-Plot 
Group : Analysis can be split based on items/levels of certain variables

Correlation Analysis

Correlation analysis :
 - Pearson R,
 - Spearman Rho, and
 - Kendall Tau
Output:
 - Complete correlation matrix table with significant sign.
 - Pairwise correlation table (r value and p value)
 - Scatter Plot (Scatter Diagram)
Group : Analysis can be split based on items/levels of certain variables

Regression Analysis

Regression Analysis:

• Simple Linear
• Multiple Linear
• Polynomial
• Response Surface
• Logistics: Logit, Probit, Multinomial, Ordinal

Features & Outputs:

• Lost data calculation
• Metode Regresi: Enter, Stepwise, Forward Selection, Backward Selection, Forward Information Criteria
• Regression Diagnostics (Normality, Homogeneity, Residual Plot and Outlier data information)
• Box-Cox Transformation
• Automatic transformation
• Automatic outlier data replacement
• Chart:
  • 2D and 3D Graphics for Response Surfaces,
  • ROC Curve, Metrik Performance, Outcome vs. Prediction
• Optimization (Maximum and Minimum)
• Confusion Matrix (Classification and Metric Tables)
• Regression Equation
• Adjusted Regression/Goodness Statistics: R ² , R ² , corrected, Correlation Coefficient; AIC, AICc, BIC, RMSE, MAE, MPE, MAPE, sMAPE
• Coef Estimation. Regression: Coefficient Value; Standard errors; t statistic, p value, Upper/Lower, VIF
• Anova: Sequential and Partial
• etc.

Student t-test and z-test

Student's t-test and z-test:

• One sample z test
• z test two independent samples
• One sample t test
• Two sample t-test (independent t-test):
  • Levene test for variance homogeneity test  
  • Welch's t-test (when the two variables are not homogeneous)
• Paired t test

Features/Outputs:

• Data layout for comparison:
  • Can compare between variables (different columns) or between levels/groups (same column, as Sample ID's)
• Alternative Hypothesis: can be one-way () or two-way (≠)
• Critical Areas & Conclusions

Non Parametric Analysis

Single Sample:

• Sign Test (Single Sample)
• Wilcoxon Signed Ranks Test (Single Sample)

Independent Sample:

• Wilcoxon Rank Sum Test
• Mann-Whitney U Test
• Kolmogorov-Smirnov Test
• Kruskal Wallis Test (One Way Anova)

Paired Sample (Dependent/Paired/Related Sample):

• Wilcoxon Signed Ranks Test (Paired Samples)
• McNemar Test
• Sign Test (Paired Samples)
• Friedman (Two Way Anova)
• Cochran Q Test

ANOVA

Experimental Design :

• Automated Creation of Layouts/Plot Layouts of various Experimental Designs
• One Way Anova/Completely Randomized Design (CRD)
• Two Way Anova/Randomized Block Design (RAK)
• Latin Square/Latin Square Design (RBSL)
• Factorial (Can be up to 3 factors, it's just that what is given further test notation is limited to only 3 factors)
• Split Plot
• Split-Spit Plot
• Strip Plot
• Mixed Design
  • Multi Location/Season/Year
  • Repeated Measure - Analyze multiple responses (Dependent Variables) at once
  • Between/Within
• Sub-Sampling (Withdrawal of sample children)

Variabel Respons:

• Analysis can be performed for multiple responses (Dependent Variables) at once

Missing Data :

• Lost data calculation (able to calculate some missing data in each Response)

Auto Transform:

• Searching for the optimal type of transformation automatically so that the results of the analysis of variance do not violate the ANOVA assumption.
• In addition to transformations, SmartstatXL can also replace outlier data values automatically, either from the original data or from the transformed data. 
• Transformation data/New data (when there is missing data/sampling) is included in the output
• Transformation Type:
  • Square Transformation
  • Square Root Transformation
  • Log Transformation
  • Inverse Square Root Transformation
  • Reciprocal Transformation
  • Inverse Square Transformation
  • ArcSin 

ANOVA table :

• The F-count value is colored red and is marked with an asterisk (*) for the influencing factor
• P-Value (Sig) and F-table values at 5% and 1% significance levels
• Value of the Coefficient of Diversity

Post Hoc :

• LSD
• Tukey’s HSD
• Scheffé’s test
• Duncan
• Student Newman Keuls (SNK)
• REGWQ
• Scott Knott
• Bonferroni
• Check
• Dunnet
• Several further tests can be used to make comparisons with controls
• Can be used for further testing of simple effects or interactions 
• Comes with Critical Values and for the Scott-Knot test included partition summaries, Bo-Max, Lambda and Chi-Square
• Table Average ± ERROR. The average value can be completed with a Standard Error, Standard Deviation, or Confidence Interval value

Display for Treatments Means (Table and Graph)

• Presentation of the average value can be in the form of tables and graphs

ANOVA Diagnostics :

• Anova Assumption Check
  • Homogeneity test (Levene test)
  • Normality Test (Shapiro Wilks, Kolmogorov-Smirnov & Liliefors, Anderson Daring, D'Agostino Pearson)
• Residual Analysis Table
  • Predicted
  • Residual
  • Studentized Residual
  • Studentized Deleted Residual
  • etc.
• Plot Resiual 
  • Normal plot
  • Histogram
  • Versus predicted values
  • Standard Deviation vs. Average value
• Box-Cox Transform (finding the optimal lambda value to choose the right type of data transformation)
• Checking outlier data and calculating missing data