Zeta Analytic Software - User Manual

Designed and Developed by Jitender Kumar

Professional Data Analysis Platform

Contents

1. Overview

Zeta Analytic Software is an offline statistical platform for researchers, scientists, hospitals, government agencies, and private organizations. It supports end-to-end workflow: import, cleaning, subset analysis, statistics, visualization, diagnostics, reporting, compliance logging, and session reproducibility.

2. System Requirements

• Windows desktop/laptop capable of running Electron applications
• Sufficient RAM for your dataset size (recommended 8 GB+ for large files)
• Input files in .xlsx, .xls, or .csv
• License file (zeta.lic) for full licensed functionality

3. Quick Start (5 Minutes)

1. Open app and click Import.
2. Select Excel/CSV file and verify table loaded.
3. Optional: apply Case Subset (e.g., patient IDs).
4. Run Summary, Charts, or Tests.
5. Use exports (CSV/Excel/PDF/Word/PNG/JPG) as required.
6. Click Save to persist full analysis session.

4. Main Screen and Toolbar

Toolbar Buttons

• Import: Load source data file.
• Summary: Open descriptive statistics and summary charts.
• Charts: Open chart builder with annotations.
• Data Correction: Open Data Correction Studio (separate window).
• Data Dictionary: Open variable dictionary editor (separate window).
• Validation Rules: Open rule preset editor and rule validation runner (separate window).
• Outlier Check: Open unusual-rows review and selective deletion tool (separate window).
• Tests: Open full statistical test environment.
• Public Health Toolkit: Open epidemiology calculators and report pack tools.
• Save / Load: Save or restore complete session state.
• CSV / Excel: Export current active dataset view.
• Undo / Redo: Reverse or reapply data operations.
• Lock Snapshot: Export tamper-evident compliance snapshot.

Status Bar

• Live status messages for operations
• License state and activation button
• Machine ID (click to copy) and application version

5. Complete Button Reference

Main Toolbar (Top)

• Import: Open file picker and load dataset.
• Summary: Open summary window for descriptive outputs and exports.
• Charts: Open chart builder window.
• Data Correction: Open issue scanner and fix workflow window.
• Data Dictionary: Open dictionary editor window.
• Validation Rules: Open rule create/load/save/run window.
• Outlier Check: Open unusual-row review/delete window.
• Tests: Open statistical tests workspace.
• Public Health Toolkit: Open epidemiology calculators.
• Save: Save current session snapshot.
• Load: Load previously saved session snapshot.
• CSV: Export current active table to CSV.
• Excel: Export current active table to XLSX.
• Activate License: Select license file and activate on this machine.

Main Screen (Left Panel + Table)

• Apply (Filter): Apply selected filter condition.
• Apply (Sort): Apply selected sort order.
• Reset All: Clear filter/sort/search and reset table view.
• Run Recommendation (Smart Analysis): Suggest suitable analysis class after checking whether current data can support it.
• Reset Smart Analysis: Clear selected variables, recommendation text, and smart-analysis launch context.
• Apply Subset: Activate IN-based subset rows for all modules.
• Clear Subset: Remove active subset and return full dataset.
• Fill Missing (Mean/Median/Mode): Impute missing cells in numeric columns.
• Winsorize Outliers: Cap extreme values to IQR bounds.
• Remove Outlier Rows: Delete rows flagged as outliers.
• Remove Duplicates: Remove duplicate records.
• Verify Audit Chain: Check hash-chain integrity of audit logs.
• Run Assumption Check: Run quick normality-oriented screening for numeric variables.
• Prev/Next: Move between table pages.

Data Correction Studio Window

• Scan Issues: Scan active dataset for cell-level issues.
• Apply All Fixes: Apply all selected auto-fixable issues.
• Manual Pending: Toggle view to focus manual-only pending issues.
• Refresh: Rebuild issue table after changes.
• Last Applied Fixes: Open correction log table window.
• Use Hint: Fill manual input with system suggestion/hint.
• Apply (manual row): Apply manually entered value for that issue.
• Help: Show manual correction guidance for that issue.
• Apply to Main Excel: Commit previewed fixes to active dataset.
• Export Separately: Export corrected copy without replacing active dataset.
• Export Correction Log: Export historical applied-fixes log.

Data Dictionary Window

• Auto Build: Generate initial dictionary from current columns.
• Save & Close: Save dictionary and close window.
• Close: Close window without additional changes.

Validation Rules Window

• Generate From Data: Auto-create starter rules using column profiles.
• Load Preset: Load saved rule preset.
• Save Preset: Save current rules as reusable preset.
• Run Validation: Validate active dataset and append rule findings to correction issues.
• Close: Close validation rules window.

Outlier Check Window

• Find Unusual Rows: Build review queue using IQR-based detection.
• Delete Checked Rows: Delete only rows selected by user.
• Close: Close outlier check window.

Summary Window

• Export CSV / Excel / PDF / PNG / JPG: Export summary table/charts in chosen format.

Charts Window

• Draw Chart: Render chart for selected variables and chart type.
• Add Annotation Text: Add movable/rotatable text annotation by placing it on chart.
• Rectangle / Circle / Free Draw: Add shape/freehand annotations.
• Delete Selected: Delete selected annotation object.
• Clear All: Remove all annotations from current chart.
• Selected text annotations support rotate-handle drag, mouse wheel/touchpad adjustment on the handle, and keyboard rotation with arrow keys.
• PNG / JPG / PDF: Export chart with annotations.

Tests Window

• Quick Tests / Advanced Tests: Switch test category.
• Run ... buttons: Execute selected statistical model/test.
• PDF / Word / PNG / JPG: Export current test output.
• Add to Report: Add selected output to the shared integrated report bundle for the current session.
• Apply Recode & Run Logistic Regression: Recode non-binary dependent values and rerun logit.
• Save (Protocol Checklist): Save checklist selections for reproducibility.
• Export Public Health Pack: Export combined public health report pack.
• Export Full Report (PDF/Word): Export all added report sections, including sections added from other Tests windows in the same session.
• Open Full View: Expand a result card for full-screen inspection.
• Print: Print current view/report output.

Public Health Toolkit Window

• Compute (Cohort): Calculate risk, RR, attributable risk.
• Compute (Case-Control): Calculate odds ratio.
• Compute (Incidence/Prevalence): Calculate frequency metrics.
• Save Checklist: Save protocol/public health checklist values.
• Export Public Health Pack (HTML): Export public-health worksheet/report package.

6. Importing Data

Supported Methods

• Toolbar Import button
• Drag and drop Excel file into central drop zone

Supported Formats

• .xlsx, .xls, .csv

7. Filtering, Sorting, Search, Pagination

Filtering

1. Select column
2. Choose operator (=, >, <, >=, <=)
3. Enter value and click Apply

Sorting

1. Select sort column
2. Select ascending/descending
3. Click Apply

Global Search and Pagination

• Use search field to match across all visible columns
• Use page-size dropdown (25/50/100 rows)
• Navigate pages with Prev/Next

8. Case Subset Analysis (IN Selection)

Use this when analysis must run on selected cases only (for example selected patient IDs).

1. In left panel, under Case Selection (IN), choose a column.
2. Enter comma-separated values, e.g., 101, 2013, 3309.
3. Click Apply Subset.

When subset is active, a label appears above table and all modules (Summary, Charts, Tests, Public Health Toolkit) run on subset rows only.

• Clear Subset returns to full dataset.

9. Smart Analysis Suggestion

Select one or two variables in Smart Analysis panel to get recommended test class. The recommendation now checks current data suitability first, so it can warn if the suggested test cannot run on the selected variables.

• Run Recommended Test opens a fresh guided Tests window so existing manual test work is not overwritten.
• Reset Smart Analysis clears current selections and removes stale recommendation context.

10. Summary Module

Outputs

• Variable-wise descriptive table
• Mean-based chart and Min-Median-Max chart
• Subset status banner (if subset is active)

Exports

• CSV, Excel
• PDF report with table and charts
• PNG/JPG chart image export

11. Charts Module

Chart Types

• Bar, Line, Pie, Histogram

Annotation Tools

• Text, Rectangle, Circle, Free Draw
• Text annotations show a visible rotate handle and selection frame for easier editing.
• Rotate text by dragging the rotate handle, using wheel/touchpad on the handle, or using arrow keys after selecting the text.
• Delete selected annotation
• Clear all annotations

Export

• PNG, JPG, PDF

12. Statistical Tests Module

Basic Tests

• Correlation
• Correlation Matrix
• Group Comparison
• One-way ANOVA
• Chi-square
• Linear Regression

Basic Tests - How to Choose (with Examples)

• Correlation: Use for relationship between two numeric variables.
  Example: Age vs Systolic BP in 300 patients.
• Correlation Matrix: Use when you want pairwise correlation across many numeric variables at once.
  Example: BMI, BP, Glucose, Cholesterol, HbA1c.
• Group Comparison: Use for quick numeric comparison by a grouping variable.
  Example: Mean Hb across treatment groups A/B/C.
• One-way ANOVA: Compare means across 3+ groups.
  Example: Recovery days across three therapy arms.
• Chi-square: Association between two categorical variables.
  Example: Smoking status vs Disease status.
• Linear Regression: Predict one numeric outcome from one predictor.
  Example: SBP predicted by BMI.

Advanced Tests

• Independent t-test
• Paired t-test
• One-sample t-test
• Multiple Linear Regression
• Logistic Regression (binary)
• Survival Analysis
• PSM / IPTW
• Diagnostics and assumption plots

Advanced Tests - How to Choose (with Examples)

• Independent t-test: Compare mean of a numeric variable between 2 independent groups.
  Example: Mean hemoglobin in Male vs Female groups.
• Paired t-test: Compare before-after measurements on same subjects.
  Example: Fasting glucose before vs after intervention in same patients.
• One-sample t-test: Compare sample mean against known reference value.
  Example: Mean BMI vs reference 25.
• Multiple Linear Regression: Predict one numeric outcome from multiple predictors.
  Example: SBP from Age, BMI, Smoking, Physical Activity.
• Logistic Regression (binary): Predict binary outcome probability.
  Example: Diabetes (Yes/No) from Age, BMI, Family History.
• Survival Analysis: Time-to-event analysis with event indicator.
  Example: Days to relapse by treatment arm with censored cases.
• PSM / IPTW: Causal adjustment for treatment vs control comparison in observational data.
  Example: Balance cohorts before estimating treatment effect on outcome.
• Diagnostics/Assumption plots: Check model/test suitability before conclusion.
  Example: Residual pattern and normality checks for regression model.

Test-by-Test Quick Guide (Inputs + Example)

• Correlation (Pearson/Spearman): Inputs = two numeric columns.
  Example: Age and Systolic BP. Use Spearman when data is skewed or ordinal-like.
• Correlation Matrix: Inputs = multiple numeric columns.
  Example: BMI, Glucose, HbA1c, LDL to see full pairwise relationship map.
• Group Comparison: Inputs = one numeric outcome + one group column.
  Example: Hemoglobin by Ward type (ICU/General).
• One-way ANOVA: Inputs = one numeric outcome + group column with 3+ levels.
  Example: Recovery days across Drug A, Drug B, Drug C.
• Chi-square: Inputs = two categorical columns.
  Example: Smoking (Yes/No) vs Disease (Yes/No).
• Linear Regression: Inputs = one numeric Y + one numeric X.
  Example: SBP predicted from BMI.
• Independent t-test: Inputs = one numeric outcome + binary group column.
  Example: Mean Hb in Male vs Female.
• Paired t-test: Inputs = two paired numeric columns from same subjects.
  Example: Glucose_Before vs Glucose_After in same patient IDs.
• One-sample t-test: Inputs = one numeric column + reference value.
  Example: Mean BMI compared against 25.
• Multiple Linear Regression: Inputs = one numeric Y + multiple X predictors.
  Example: SBP from Age, BMI, Smoking, Physical Activity.
• Logistic Regression (Binary): Inputs = one binary outcome + predictors.
  Example: Diabetes (0/1) from Age, BMI, Family History.
• Survival Analysis: Inputs = time variable + event variable + optional group variable.
  Example: Follow-up days + Event(0/1) + Treatment Group.
• PSM / IPTW: Inputs = treatment indicator + outcome + confounders.
  Example: Treatment(0/1), Mortality(0/1), Age/BMI/Comorbidity predictors.
• Assumption Diagnostics: Inputs = generated automatically from selected model variables.
  Example: Residual normality warning indicates parametric assumptions may be weak.

Reproducibility & Exports

• Add any output to report
• Report sections are shared across Tests windows in the same working session
• Export section: PDF/Word/PNG/JPG
• Export full report: PDF/Word
• Protocol checklist + Public Health Pack export

13. Public Health Toolkit

Dedicated workspace for epidemiology-oriented calculations and reporting.

Included Tools

• Cohort metrics: risk in exposed/unexposed, RR, attributable risk
• Case-control metric: odds ratio
• Incidence and prevalence calculator
• Protocol checklist capture
• Public Health Report Pack export (HTML)

14. Data Quality Tools

Missing Data Handling

• Fill by Mean
• Fill by Median
• Fill by Mode

Outlier Handling (IQR)

• Winsorize outliers to bounds
• Remove outlier rows

Duplicates

• Remove duplicate records

Quality Profile

• Rows, columns, numeric columns, missing cells, outlier columns

Data Correction Studio (Top Toolbar)

• Scan issues at cell level
• Apply selected fixable issues with preview before applying
• Manual correction for non-auto-fixable issues
• Last Applied Fixes window with correction history

Validation Rules (Top Toolbar)

• Create reusable column rules (type, required, min/max, allowed values)
• Save/load preset for repeated organizational workflows
• Run rule validation to append rule-based issues into Data Correction workflow

Outlier Check (Top Toolbar)

• Find unusual rows using IQR rule
• Shows row number + ID column/value for easy verification
• User decides keep/remove; tool does not force deletion

Data Dictionary (Top Toolbar)

• Define variable label, type, allowed values, and missing policy
• Improves consistency across projects and team users
• Can be auto-built then edited and saved

Assumption Check (Left Panel)

• Quick numeric-variable screening for likely non-normal patterns
• Helps users choose parametric vs robust/non-parametric tests
• Displayed in Tests module as Assumption Diagnostics summary

15. Audit Trail and Locked Snapshot

Audit Trail

• Every major action is logged with timestamp and hash chain
• Verify Audit Chain checks integrity

Locked Snapshot

• Exports compliance JSON including methods, logs, subset, and data hash
• Use when submitting analysis for review/audit

16. Session Save/Load and Reproducibility

Save Includes

• Original and current dataset state
• Analysis log
• Methods notes
• Validation rules
• Dataset dictionary
• Assumption diagnostics snapshot
• Report sections from the shared session report bundle
• Chart annotations

Load Behavior

• Restores prior analysis environment
• Allows continuation of reporting and exports

Auto Recovery

• If app closes unexpectedly, software can offer restore of last unsaved workspace on next start
• Recommended to accept restore, verify dataset, then save session explicitly

17. License Activation and Status

How to Buy License

1. Open official website and go to Pricing.
2. Click Buy 30 Days, Buy 1 Year, or Buy Lifetime based on your plan.
3. Your email draft opens automatically with required fields prefilled.
4. Click the Machine ID in the software status bar to copy it, then paste it into your request.
5. Send the email to jtndr.kmr01@gmail.com.
6. After payment confirmation, vendor sends machine-bound zeta.lic file.
7. Use Activate License in software to apply the file.

18. Keyboard Shortcuts

• Ctrl + Z: Undo
• Ctrl + Y: Redo
• Ctrl + F: Focus global search

19. Recommended Analysis Workflow

1. Import and verify headers/types
2. Apply data-quality checks and cleaning
3. Apply subset if protocol requires case-based analysis
4. Run descriptive summary first
5. Run inferential tests and diagnostics
6. Add key outputs to report
7. Save session and export locked snapshot for compliance
8. Export final report pack

20. Troubleshooting

Import Issues

• Ensure file has valid header row
• Check file is not empty
• Retry with clean CSV/Excel format

No Test Output

• Confirm variable types and non-missing values
• Check subset is not too restrictive
• Ensure minimum sample conditions for selected test

Report Export Blocked

• Complete required protocol checklist items in Tests module

License Not Valid

• Confirm license belongs to current machine ID
• Check license expiry and activation status

21. Support

For technical support, licensing, deployment assistance, and training:
jtndr.kmr01@gmail.com