No-Code Python Compatible Statistics
The output results are comparable to those obtained by running major Python libraries, ensuring reliability for academic use.
New Release Campaign: 50% off the regular price of $400 for a perpetual license
For higher-performance Machine Learning models
PyStat-Base covers the most commonly used statistical analysis techniques: tabulation, basic statistics, hypothesis testing, multivariate analysis, graphing, etc. You will be able to master practical statistics in no time
In addition to the functions of Free and Base, PyStat-Pro has enhanced functions such as nonparametric tests, multiple comparison tests, predictive models, and machine learning to meet more advanced needs.
PyStat-Premium provides all the features of Pystat in addition to the features of Free, Base, and Pro, and can meet all your statistical needs. Especially, time series analysis, survival analysis, and meta-analysis are enhanced.
Python-compatible statistical analysis software
PyStat is a statistical analysis software that guarantees equivalent calculation results to those using Python’s statistical and machine learning libraries. There is no need to worry about security as all processing is done within your computer.
Note: Pystat is a Python-compatible statistical analysis software, but does not provide a Python virtual environment or related libraries (Python coding and execution of arbitrary Python code are not possible).
Generative AI supports you.
The OpenAI’s latest LLM will explain how to interpret the results, so you can avoid misusing statistical methods.
Transparent licensing policy
Many commercial statistical analysis software products offer pricing for businesses, faculty, and students. However, the distinction between these statuses is unclear and does not reflect the reality of joint research between companies and universities, or collaboration between faculty and students. PyStat offers academic pricing levels for all users.
Comparison Table
Find the right product with the features you need.
| Description | Free |
Base |
Pro |
Premium |
|
| AI Asistance | |||||
| Automatic Consideration | PyStat will use ChatGPT to analyze the results calculated by each analysis function. | ✓ |
✓ |
✓ |
|
| Basic Statistcs | |||||
| Descriptive Statistics | Basic statistics such as the mean, variance, and median can be calculated all at once. Statistics can also be calculated for each group. | ✓ |
✓ |
✓ |
✓ |
| Correlation Coefficient | It is possible to perform Pearson product-moment correlation coefficients, rank correlation coefficients, and tests for non-correlation. | ✓ |
✓ |
✓ |
|
| Intraclass Correlation Coefficient | Intraclass correlation coefficients and 95% confidence intervals can be calculated. | ✓ |
|||
| Hypothesis Tests | |||||
| Checking Normality | Three methods are possible: – Creating a normal Q-Q plot – Shapiro-Wilk test – K-S test | ✓ |
✓ |
✓ |
✓ |
| Test of Homogeneity of Variance | Three test methods are available: – F test – Bartlett test – Levene test | ✓ |
✓ |
||
| Parametric Tests | |||||
| Unpaired t-test | Calculate statistics, p-values, and effect sizes for unpaired t-tests. | ✓ |
✓ |
✓ |
✓ |
| Welch’s t-test | Calculate Welch’s t-test statistics, p-values, and effect sizes. | ✓ |
✓ |
✓ |
✓ |
| Paired t-test | Calculate statistics, p-values, and effect sizes for paired t-tests. | ✓ |
✓ |
✓ |
✓ |
| One-way ANOVA | Perform a one-way ANOVA and create an ANOVA table. | ✓ |
✓ |
✓ |
✓ |
| Repeated Measures ANOVA | Conduct one-way ANOVA with replicates. | ✓ |
✓ |
||
| Factorial ANOVA | Perform multi-way ANOVA, such as two-way or three-way ANOVA, and create ANOVA tables. | ✓ |
✓ |
✓ |
✓ |
| Non-parametric Tests | |||||
| Mann–Whitney U test | Calculate the Mann-Whitney U test statistic, p-value, and effect size. | ✓ |
|||
| Brunner-Munzel test | Calculate the Brunner-Munchel test statistic and p-value. | ✓ |
|||
| Wilcoxon rank sum test | Calculate Wilcoxon signed-rank test statistics, p-values, and effect sizes. | ✓ |
|||
| Kruskal-Wallis test | Calculate the Kruskal-Wallis test statistic and p-value. | ✓ |
|||
| Friedman’s test | Calculate Friedman test statistics, p-values, and effect sizes. | ✓ |
|||
| Multiple Comparison | |||||
| Tukey-Kramer’s test | Calculate the p-value for the Tukey-Kramer method. | ✓ |
✓ |
||
| Steel-Dwass test | Calculate the p-value for the Steele-Dwas test. | ✓ |
✓ |
||
| Dunnett’s test | Calculate the p-value for the Dunnett test. *Not available on the macOS version. | ✓ |
✓ |
||
| Bonferroni correction | Bonferroni-type corrections are performed on significance levels and p-values. Sidak and Holm methods can also be performed. | ✓ |
✓ |
||
| Multivariate Analysis | |||||
| Discriminant Analysis | Perform discriminant analysis to calculate contribution ratios, perform classification, and create graphs. | ✓ |
✓ |
✓ |
|
| Principal Component Analysis | Perform principal component analysis to obtain contribution ratios, principal component loadings, and create graphs. | ✓ |
✓ |
✓ |
|
| Factor Analysis | Perform factor analysis to generate eigenvalues, factor loadings, and biplots. | ✓ |
✓ |
✓ |
|
| Structured Equation Modeling | Define the model formula and perform structural equation modeling. The results are displayed as a path diagram. | ✓ |
✓ |
||
| Regression | |||||
| Multiple Regression | Perform multiple regression analysis to calculate the coefficient of determination, partial regression coefficient, and VIF. It is also possible to evaluate the model and calculate predicted values. | ✓ |
✓ |
✓ |
|
| Logistic Regression Analysis | Performs logistic regression analysis to calculate log-likelihoods, partial regression coefficients, and adjusted odds ratios. | ✓ |
✓ |
✓ |
✓ |
| Multinomial Logistic Regression Analysis | Runs a multinomial logistic regression and calculate the log-likelihood and explanatory variable statistics. | ✓ |
✓ |
||
| Probit Analysis | Performs probit analysis to calculate log likelihood, partial regression coefficients, and marginal effects. Also can evaluate the model and calculate predicted values. | ✓ |
|||
| Generalized Linear Mixed Model | Analysis can be performed using generalized linear mixed models. | ✓ |
✓ |
||
| Quantification Theory | |||||
| Type Ⅱ | Executes quantification type II to calculate correlation ratios, ranges, and partial correlation coefficients. | ✓ |
|||
| Type Ⅲ | Quantification Type III is performed to calculate cumulative scores, category scores, and sample scores, and graphs are created. | ✓ |
|||
| Machine Learning | |||||
| Decision Tree | It is possible to perform classification and regression problems using decision trees. | ✓ |
✓ |
✓ |
✓ |
| Random Forest | Classification and regression problems can be solved using random forests. | ✓ |
✓ |
||
| Gradient Boosting Decision Tree | It can perform classification and regression problems using gradient boosting decision tree | ✓ |
✓ |
||
| Support Vector Machine | It can perform classification and regression problems using support vector machine | ✓ |
✓ |
||
| Time Series Analysis | |||||
| Basic Analysis | |||||
| Moving Average | It can perform classification and regression problems using gradient boosting decision trees. | ✓ |
✓ |
✓ |
✓ |
| Autocorrelation Coefficient | Calculate the autocorrelation coefficient and partial autocorrelation coefficient and create a correlogram. | ✓ |
✓ |
✓ |
✓ |
| Augmented Dickey-Fuller test | It is possible to perform the ADF test, which is a test for unit root processes. | ✓ |
✓ |
||
| Models | |||||
| ARIMA | Create models and make predictions using the ARIMA model. Parameters can also be calculated automatically. | ✓ |
✓ |
||
| GARCH | Using the GARCH model, it is possible to create models and calculate volatility and VaR. | ✓ |
|||
| VAR | Use VAR models to create models, make forecasts, and analyze causality using impulse response functions and Granger causality tests. | ✓ |
|||
| Prophet | Prophet can be used to make predictions on time series data. | ✓ |
|||
| Graphs | |||||
| Line Chart | Create a line graph. | ✓ |
✓ |
✓ |
✓ |
| Scatter Plot | Create scatter and bubble charts. | ✓ |
✓ |
✓ |
✓ |
| Scatterplot Matrix | Create a scatter plot matrix for multivariate data. | ✓ |
✓ |
✓ |
✓ |
| Histogram | Create a histogram. It is possible to create a histogram by overlaying two data. | ✓ |
✓ |
✓ |
|
| Distribution Plot | Creates four graphs: – Box plot – Violin plot – Swarm plot – Split plot | ✓ |
✓ |
✓ |
|
| Error bar | Create error bars. | ✓ |
✓ |
✓ |
✓ |
| Other methods | |||||
| Crosstab | Create cross-tabulation tables and perform chi-square tests, exact tests, odds ratios, risk ratios, and correspondence analysis. | ✓ |
✓ |
✓ |
|
| Association Analysis | Association analysis is performed based on various rules. Network diagrams can also be created. | ✓ |
✓ |
||
| Ppropensity Score Matching | Based on the propensity scores, greedy matching or optimal matching is performed. It is also possible to check the balance of the matching results. | ✓ |
|||
| Survival Time Analysis | |||||
| Survival Curve | It is possible to create survival curves using the Kaplan-Meier method, and perform log-rank tests and generalized Wilcoxon tests. | ✓ |
|||
| Cox Proportional Hazards Regression | Hazard ratios can be calculated and tested for survival time data. | ✓ |
|||
| Meta-Analysis | |||||
| Meta-Analysis of Proportions | Perform meta-analysis on categorical data. Forest and funnel plots can also be created. | ✓ |
|||
| Meta-Analysis of means | Perform meta-analysis on continuous data. Forest and funnel plots can also be created. | ✓ |
|||
| Generalized inverse variance method | Meta-analysis is performed using the generalized inverse variance method. Forest plots and funnel plots can also be created. | ✓ |
Let’s download PyStat-Free!
Basic statistics features are free, forever.