nettracer3d 0.9.8__py3-none-any.whl → 1.1.5__py3-none-any.whl

nettracer3d/stats.py

@@ -0,0 +1,861 @@
+import sys
+import pandas as pd
+import numpy as np
+from PyQt6.QtWidgets import (QApplication, QMainWindow, QWidget, QVBoxLayout, 
+                            QHBoxLayout, QLabel, QTableWidget, QTableWidgetItem,
+                            QPushButton, QComboBox, QTextEdit, QSplitter, 
+                            QListWidget, QListWidgetItem, QFrame, QMessageBox,
+                            QHeaderView, QAbstractItemView, QCheckBox)
+from PyQt6.QtCore import Qt, QMimeData
+from PyQt6.QtGui import QDragEnterEvent, QDropEvent, QFont, QDrag
+from scipy import stats
+import os
+
+class DragDropListWidget(QListWidget):
+    """Custom QListWidget that accepts drag and drop of column names with remove buttons"""
+    
+    def __init__(self, parent=None):
+        super().__init__(parent)
+        self.setAcceptDrops(True)
+        self.setDragDropMode(QAbstractItemView.DragDropMode.InternalMove)
+        self.setDefaultDropAction(Qt.DropAction.MoveAction)
+        
+    def dragEnterEvent(self, event: QDragEnterEvent):
+        if event.mimeData().hasText():
+            event.acceptProposedAction()
+        else:
+            event.ignore()
+            
+    def dropEvent(self, event: QDropEvent):
+        if event.mimeData().hasText():
+            column_name = event.mimeData().text()
+            # Check if column already exists
+            for i in range(self.count()):
+                widget = self.itemWidget(self.item(i))
+                if widget and widget.column_name == column_name:
+                    return  # Column already exists
+            
+            self.add_column_item(column_name)
+            event.acceptProposedAction()
+        else:
+            event.ignore()
+    
+    def add_column_item(self, column_name):
+        """Add a new column item with remove button"""
+        from PyQt6.QtWidgets import QWidget, QHBoxLayout, QLabel, QPushButton
+        
+        # Create list item
+        item = QListWidgetItem()
+        self.addItem(item)
+        
+        # Create custom widget for the item
+        widget = QWidget()
+        widget.column_name = column_name  # Store column name for reference
+        layout = QHBoxLayout(widget)
+        layout.setContentsMargins(5, 2, 5, 2)
+        
+        # Column name label
+        label = QLabel(column_name)
+        label.setStyleSheet("QLabel { color: #333; }")
+        layout.addWidget(label)
+        
+        # Remove button
+        remove_btn = QPushButton("×")
+        remove_btn.setMaximumSize(20, 20)
+        remove_btn.setStyleSheet("""
+            QPushButton {
+                background-color: #ff4444;
+                color: white;
+                border: none;
+                border-radius: 10px;
+                font-weight: bold;
+                font-size: 12px;
+            }
+            QPushButton:hover {
+                background-color: #cc3333;
+            }
+            QPushButton:pressed {
+                background-color: #aa2222;
+            }
+        """)
+        
+        # Connect remove button to removal function
+        remove_btn.clicked.connect(lambda: self.remove_column_item(item))
+        layout.addWidget(remove_btn)
+        
+        # Set the custom widget for this item
+        item.setSizeHint(widget.sizeHint())
+        self.setItemWidget(item, widget)
+    
+    def remove_column_item(self, item):
+        """Remove a column item from the list"""
+        row = self.row(item)
+        if row >= 0:
+            self.takeItem(row)
+    
+    def get_selected_columns(self):
+        """Get list of all selected column names"""
+        columns = []
+        for i in range(self.count()):
+            widget = self.itemWidget(self.item(i))
+            if widget and hasattr(widget, 'column_name'):
+                columns.append(widget.column_name)
+        return columns
+
+class DragDropTableWidget(QTableWidget):
+    """Custom QTableWidget that allows dragging column headers"""
+    
+    def __init__(self, parent=None):
+        super().__init__(parent)
+        self.setDragEnabled(True)
+        self.setDragDropMode(QAbstractItemView.DragDropMode.DragOnly)
+        self.setDefaultDropAction(Qt.DropAction.CopyAction)
+        
+    def startDrag(self, supportedActions):
+        if self.currentColumn() >= 0:
+            # Create drag data with column index and header
+            drag = QDrag(self)
+            mimeData = QMimeData()
+            
+            # Store column index and header text
+            col_idx = self.currentColumn()
+            header_text = self.horizontalHeaderItem(col_idx).text() if self.horizontalHeaderItem(col_idx) else f"Column_{col_idx}"
+            
+            # Just store the header text (simpler than the reference)
+            mimeData.setText(header_text)
+            drag.setMimeData(mimeData)
+            
+            # Create drag pixmap for visual feedback
+            from PyQt6.QtGui import QPixmap, QPainter
+            pixmap = QPixmap(150, 30)
+            pixmap.fill(Qt.GlobalColor.lightGray)
+            painter = QPainter(pixmap)
+            painter.drawText(pixmap.rect(), Qt.AlignmentFlag.AlignCenter, header_text)
+            painter.end()
+            drag.setPixmap(pixmap)
+            
+            drag.exec(Qt.DropAction.CopyAction)
+
+class FileDropWidget(QWidget):
+    """Widget that accepts file drops"""
+    
+    def __init__(self, parent=None):
+        super().__init__(parent)
+        self.setAcceptDrops(True)
+        self.parent_window = parent
+        
+        # Setup UI
+        layout = QVBoxLayout()
+        self.label = QLabel("Drop .xlsx or .csv files here")
+        self.label.setAlignment(Qt.AlignmentFlag.AlignCenter)
+        self.label.setStyleSheet("""
+            QLabel {
+                border: 2px dashed #aaa;
+                border-radius: 10px;
+                padding: 20px;
+                background-color: #f9f9f9;
+                font-size: 14px;
+                color: #666;
+            }
+        """)
+        layout.addWidget(self.label)
+        self.setLayout(layout)
+        
+    def dragEnterEvent(self, event: QDragEnterEvent):
+        if event.mimeData().hasUrls():
+            urls = event.mimeData().urls()
+            for url in urls:
+                if url.isLocalFile():
+                    file_path = url.toLocalFile()
+                    if file_path.lower().endswith(('.xlsx', '.csv')):
+                        event.acceptProposedAction()
+                        return
+        event.ignore()
+        
+    def dropEvent(self, event: QDropEvent):
+        if event.mimeData().hasUrls():
+            urls = event.mimeData().urls()
+            for url in urls:
+                if url.isLocalFile():
+                    file_path = url.toLocalFile()
+                    if file_path.lower().endswith(('.xlsx', '.csv')):
+                        self.parent_window.load_file(file_path)
+                        event.acceptProposedAction()
+                        return
+        event.ignore()
+
+class StatisticalTestGUI(QMainWindow):
+    def __init__(self):
+        super().__init__()
+        self.current_dataframe = None
+        self.init_ui()
+        
+    def init_ui(self):
+        self.setWindowTitle("Statistical Testing GUI")
+        self.setGeometry(100, 100, 1200, 800)
+        
+        # Central widget
+        central_widget = QWidget()
+        self.setCentralWidget(central_widget)
+        
+        # Main layout
+        main_layout = QHBoxLayout(central_widget)
+        
+        # Create splitter for resizable sections
+        splitter = QSplitter(Qt.Orientation.Horizontal)
+        main_layout.addWidget(splitter)
+        
+        # Left panel - File staging and data display
+        left_panel = QWidget()
+        left_layout = QVBoxLayout(left_panel)
+        
+        # File drop area
+        file_drop_label = QLabel("Data Staging Area")
+        file_drop_label.setFont(QFont("Arial", 12, QFont.Weight.Bold))
+        left_layout.addWidget(file_drop_label)
+        
+        self.file_drop_widget = FileDropWidget(self)
+        self.file_drop_widget.setMaximumHeight(100)
+        left_layout.addWidget(self.file_drop_widget)
+        
+        # Data display table
+        data_display_label = QLabel("Data Display")
+        data_display_label.setFont(QFont("Arial", 12, QFont.Weight.Bold))
+        left_layout.addWidget(data_display_label)
+        
+        self.data_table = DragDropTableWidget()
+        self.data_table.setAlternatingRowColors(True)
+        left_layout.addWidget(self.data_table)
+        
+        splitter.addWidget(left_panel)
+        
+        # Right panel - Column selection and testing
+        right_panel = QWidget()
+        right_layout = QVBoxLayout(right_panel)
+        
+        # Column selection area
+        cocking_label = QLabel("Dataset Selection Area")
+        cocking_label.setFont(QFont("Arial", 12, QFont.Weight.Bold))
+        right_layout.addWidget(cocking_label)
+        
+        cocking_instruction = QLabel("Drag column headers here to select datasets for comparison")
+        cocking_instruction.setStyleSheet("color: #666; font-style: italic;")
+        right_layout.addWidget(cocking_instruction)
+        
+        self.column_list = DragDropListWidget()
+        self.column_list.setMaximumHeight(150)
+        right_layout.addWidget(self.column_list)
+        
+        # Clear columns button
+        clear_button = QPushButton("Clear Selected Columns")
+        clear_button.clicked.connect(self.clear_columns)
+        right_layout.addWidget(clear_button)
+        
+        # Test selection
+        test_label = QLabel("Statistical Test Selection")
+        test_label.setFont(QFont("Arial", 12, QFont.Weight.Bold))
+        right_layout.addWidget(test_label)
+        
+        self.test_combo = QComboBox()
+        self.test_combo.addItems([
+            "Student's t-test (independent)",
+            "Student's t-test (paired)",
+            "Welch's t-test (independent)",
+            "Welch's t-test (paired)",
+            "One-way ANOVA",
+            "Mann-Whitney U test",
+            "Correlation analysis (Pearson)",
+            "Normality test (Shapiro-Wilk)",
+            "Chi-square test of independence"
+        ])
+        self.test_combo.currentTextChanged.connect(self.update_test_info)
+        right_layout.addWidget(self.test_combo)
+        
+        # Test information label
+        self.test_info_label = QLabel()
+        self.test_info_label.setStyleSheet("color: #666; font-style: italic; padding: 5px;")
+        self.test_info_label.setWordWrap(True)
+        right_layout.addWidget(self.test_info_label)
+        
+        # Update test info initially
+        self.update_test_info()
+        
+        # Execute button
+        self.execute_button = QPushButton("Execute Statistical Test")
+        self.execute_button.clicked.connect(self.execute_test)
+        self.execute_button.setStyleSheet("""
+            QPushButton {
+                background-color: #4CAF50;
+                color: white;
+                border: none;
+                padding: 10px;
+                font-size: 14px;
+                font-weight: bold;
+                border-radius: 5px;
+            }
+            QPushButton:hover {
+                background-color: #45a049;
+            }
+            QPushButton:pressed {
+                background-color: #3e8e41;
+            }
+        """)
+        right_layout.addWidget(self.execute_button)
+        
+        # Output area
+        output_label = QLabel("Test Results")
+        output_label.setFont(QFont("Arial", 12, QFont.Weight.Bold))
+        right_layout.addWidget(output_label)
+        
+        self.output_text = QTextEdit()
+        self.output_text.setReadOnly(True)
+        self.output_text.setMaximumHeight(300)
+        right_layout.addWidget(self.output_text)
+        
+        splitter.addWidget(right_panel)
+        
+        # Set splitter proportions
+        splitter.setSizes([700, 500])
+    
+    def update_test_info(self):
+        """Update the test information label based on selected test"""
+        test_type = self.test_combo.currentText()
+        info_text = ""
+        
+        if "t-test (independent)" in test_type:
+            info_text = "Requires: 2 columns (independent groups)\nCompares means of two separate groups"
+        elif "t-test (paired)" in test_type:
+            info_text = "Requires: 2 columns (same subjects measured twice)\nCompares paired observations"
+        elif "One-way ANOVA" in test_type:
+            info_text = "Requires: 2+ columns\nCompares means across multiple groups"
+        elif "Mann-Whitney U" in test_type:
+            info_text = "Requires: 2 columns (independent groups)\nNon-parametric alternative to independent t-test"
+        elif "Correlation" in test_type:
+            info_text = "Requires: 2 columns\nMeasures linear relationship between variables"
+        elif "Normality test" in test_type:
+            info_text = "Requires: 1+ columns\nTests if data follows normal distribution"
+        elif "Chi-square" in test_type:
+            info_text = "Requires: 2 columns (categorical)\nTests independence between categorical variables"
+        
+        self.test_info_label.setText(info_text)
+        
+    def load_file(self, file_path):
+        """Load CSV or Excel file into pandas DataFrame and display in table"""
+        try:
+            if file_path.lower().endswith('.csv'):
+                df = pd.read_csv(file_path)
+            elif file_path.lower().endswith('.xlsx'):
+                df = pd.read_excel(file_path)
+            else:
+                self.show_error("Unsupported file format. Please use .csv or .xlsx files.")
+                return
+                
+            self.current_dataframe = df
+            self.populate_table(df)
+            
+            # Update file drop label
+            filename = os.path.basename(file_path)
+            self.file_drop_widget.label.setText(f"Loaded: {filename}")
+            self.file_drop_widget.label.setStyleSheet("""
+                QLabel {
+                    border: 2px solid #4CAF50;
+                    border-radius: 10px;
+                    padding: 20px;
+                    background-color: #e8f5e8;
+                    font-size: 14px;
+                    color: #2e7d2e;
+                }
+            """)
+            
+            self.output_text.append(f"Successfully loaded {filename} with {df.shape[0]} rows and {df.shape[1]} columns.")
+            
+        except Exception as e:
+            self.show_error(f"Error loading file: {str(e)}")
+            
+    def populate_table(self, df):
+        """Populate the table widget with DataFrame data"""
+        self.data_table.setRowCount(min(df.shape[0], 1000))  # Limit display to 1000 rows
+        self.data_table.setColumnCount(df.shape[1])
+        
+        # Set headers
+        self.data_table.setHorizontalHeaderLabels(df.columns.tolist())
+        
+        # Populate data (limit to first 1000 rows for performance)
+        for i in range(min(df.shape[0], 1000)):
+            for j in range(df.shape[1]):
+                item = QTableWidgetItem(str(df.iloc[i, j]))
+                self.data_table.setItem(i, j, item)
+        
+        # Resize columns to content
+        self.data_table.horizontalHeader().setStretchLastSection(True)
+        self.data_table.resizeColumnsToContents()
+        
+        if df.shape[0] > 1000:
+            self.output_text.append(f"Note: Displaying first 1000 rows of {df.shape[0]} total rows.")
+            
+    def clear_columns(self):
+        """Clear all selected columns"""
+        self.column_list.clear()
+        
+    def get_numeric_column_data(self, column_name):
+        """Extract numeric data from a column, excluding NaN and non-numeric values"""
+        if self.current_dataframe is None:
+            return None
+            
+        if column_name not in self.current_dataframe.columns:
+            return None
+            
+        column_data = self.current_dataframe[column_name]
+        
+        # Convert to numeric, coercing errors to NaN
+        numeric_data = pd.to_numeric(column_data, errors='coerce')
+        
+        # Remove NaN values
+        clean_data = numeric_data.dropna()
+        
+        return clean_data.values
+    
+    def get_categorical_column_data(self, column_name):
+        """Extract categorical data from a column, excluding NaN values"""
+        if self.current_dataframe is None:
+            return None
+            
+        if column_name not in self.current_dataframe.columns:
+            return None
+            
+        column_data = self.current_dataframe[column_name]
+        
+        # Remove NaN values
+        clean_data = column_data.dropna()
+        
+        return clean_data.values
+        
+    def execute_test(self):
+        """Execute the selected statistical test"""
+        if self.current_dataframe is None:
+            self.show_error("Please load a dataset first.")
+            return
+            
+        # Get selected columns
+        selected_columns = self.column_list.get_selected_columns()
+            
+        if len(selected_columns) == 0:
+            self.show_error("Please select at least one column for testing.")
+            return
+            
+        # Get test type
+        test_type = self.test_combo.currentText()
+        
+        try:
+            if "Student's t-test (independent)" in test_type:
+                self.execute_ttest(selected_columns, paired=False, equal_var=True)
+            elif "Student's t-test (paired)" in test_type:
+                self.execute_ttest(selected_columns, paired=True, equal_var=True)
+            elif "Welch's t-test (independent)" in test_type:
+                self.execute_ttest(selected_columns, paired=False, equal_var=False)
+            elif "Welch's t-test (paired)" in test_type:
+                self.execute_ttest(selected_columns, paired=True, equal_var=False)
+            elif "One-way ANOVA" in test_type:
+                self.execute_anova(selected_columns)
+            elif "Mann-Whitney U test" in test_type:
+                self.execute_mannwhitney(selected_columns)
+            elif "Correlation analysis" in test_type:
+                self.execute_correlation(selected_columns)
+            elif "Normality test" in test_type:
+                self.execute_normality_test(selected_columns)
+            elif "Chi-square test" in test_type:
+                self.execute_chisquare(selected_columns)
+        except Exception as e:
+            self.show_error(f"Error executing test: {str(e)}")
+            
+    def execute_ttest(self, selected_columns, paired=False, equal_var=True):
+        """Execute t-test (Student's or Welch's, paired or independent)"""
+        if len(selected_columns) != 2:
+            self.show_error(f"t-test requires exactly 2 columns. You have selected {len(selected_columns)} columns.")
+            return
+            
+        # Extract numeric data from columns
+        data1 = self.get_numeric_column_data(selected_columns[0])
+        data2 = self.get_numeric_column_data(selected_columns[1])
+        
+        if data1 is None or len(data1) == 0:
+            self.show_error(f"Column '{selected_columns[0]}' contains no numeric data.")
+            return
+        if data2 is None or len(data2) == 0:
+            self.show_error(f"Column '{selected_columns[1]}' contains no numeric data.")
+            return
+            
+        if paired and len(data1) != len(data2):
+            self.show_error("Paired t-test requires equal sample sizes.")
+            return
+            
+        # Perform appropriate t-test
+        if paired:
+            statistic, p_value = stats.ttest_rel(data1, data2)
+        else:
+            statistic, p_value = stats.ttest_ind(data1, data2, equal_var=equal_var)
+        
+        # Display results
+        self.output_text.clear()
+        test_name = "WELCH'S" if not equal_var else "STUDENT'S"
+        test_type_desc = "PAIRED" if paired else "INDEPENDENT"
+        
+        self.output_text.append("=" * 50)
+        self.output_text.append(f"{test_name} T-TEST ({test_type_desc}) RESULTS")
+        self.output_text.append("=" * 50)
+        self.output_text.append(f"Group 1: {selected_columns[0]}")
+        self.output_text.append(f"  Sample size (n₁): {len(data1)}")
+        self.output_text.append(f"  Mean: {np.mean(data1):.4f}")
+        self.output_text.append(f"  Std Dev: {np.std(data1, ddof=1):.4f}")
+        self.output_text.append("")
+        self.output_text.append(f"Group 2: {selected_columns[1]}")
+        self.output_text.append(f"  Sample size (n₂): {len(data2)}")
+        self.output_text.append(f"  Mean: {np.mean(data2):.4f}")
+        self.output_text.append(f"  Std Dev: {np.std(data2, ddof=1):.4f}")
+        self.output_text.append("")
+        self.output_text.append("TEST STATISTICS:")
+        self.output_text.append(f"  t-statistic: {statistic:.6f}")
+        self.output_text.append(f"  p-value: {p_value:.6f}")
+        
+        if paired:
+            df = len(data1) - 1
+        else:
+            df = len(data1) + len(data2) - 2
+        self.output_text.append(f"  Degrees of freedom: {df}")
+        self.output_text.append("")
+        
+        # Interpretation
+        alpha = 0.05
+        if p_value < alpha:
+            self.output_text.append(f"RESULT: Significant difference (p < {alpha})")
+        else:
+            self.output_text.append(f"RESULT: No significant difference (p ≥ {alpha})")
+            
+    def execute_anova(self, selected_columns):
+        """Execute one-way ANOVA"""
+        if len(selected_columns) < 2:
+            self.show_error(f"One-way ANOVA requires at least 2 columns. You have selected {len(selected_columns)} columns.")
+            return
+            
+        # Extract numeric data from all columns
+        datasets = []
+        dataset_info = []
+        
+        for col_name in selected_columns:
+            data = self.get_numeric_column_data(col_name)
+            if data is None or len(data) == 0:
+                self.show_error(f"Column '{col_name}' contains no numeric data.")
+                return
+            datasets.append(data)
+            dataset_info.append({
+                'name': col_name,
+                'n': len(data),
+                'data': data
+            })
+            
+        # Perform one-way ANOVA
+        statistic, p_value = stats.f_oneway(*datasets)
+        
+        # Display results
+        self.output_text.clear()
+        self.output_text.append("=" * 50)
+        self.output_text.append("ONE-WAY ANOVA RESULTS")
+        self.output_text.append("=" * 50)
+        
+        for i, group in enumerate(dataset_info):
+            self.output_text.append(f"Group {i+1}: {group['name']}")
+            self.output_text.append(f"  Sample size (n): {group['n']}")
+            self.output_text.append(f"  Mean: {np.mean(group['data']):.4f}")
+            self.output_text.append(f"  Std Dev: {np.std(group['data'], ddof=1):.4f}")
+            self.output_text.append("")
+            
+        self.output_text.append("TEST STATISTICS:")
+        self.output_text.append(f"  F-statistic: {statistic:.6f}")
+        self.output_text.append(f"  p-value: {p_value:.6f}")
+        
+        # Degrees of freedom
+        k = len(datasets)  # number of groups
+        N = sum(len(data) for data in datasets)  # total sample size
+        df_between = k - 1
+        df_within = N - k
+        
+        self.output_text.append(f"  Degrees of freedom (between): {df_between}")
+        self.output_text.append(f"  Degrees of freedom (within): {df_within}")
+        self.output_text.append("")
+        
+        # Interpretation
+        alpha = 0.05
+        if p_value < alpha:
+            self.output_text.append(f"RESULT: Significant difference between groups (p < {alpha})")
+        else:
+            self.output_text.append(f"RESULT: No significant difference between groups (p ≥ {alpha})")
+            
+    def execute_mannwhitney(self, selected_columns):
+        """Execute Mann-Whitney U test"""
+        if len(selected_columns) != 2:
+            self.show_error(f"Mann-Whitney U test requires exactly 2 columns. You have selected {len(selected_columns)} columns.")
+            return
+            
+        # Extract numeric data from columns
+        data1 = self.get_numeric_column_data(selected_columns[0])
+        data2 = self.get_numeric_column_data(selected_columns[1])
+        
+        if data1 is None or len(data1) == 0:
+            self.show_error(f"Column '{selected_columns[0]}' contains no numeric data.")
+            return
+        if data2 is None or len(data2) == 0:
+            self.show_error(f"Column '{selected_columns[1]}' contains no numeric data.")
+            return
+            
+        # Perform Mann-Whitney U test
+        statistic, p_value = stats.mannwhitneyu(data1, data2, alternative='two-sided')
+        
+        # Display results
+        self.output_text.clear()
+        self.output_text.append("=" * 50)
+        self.output_text.append("MANN-WHITNEY U TEST RESULTS")
+        self.output_text.append("=" * 50)
+        self.output_text.append(f"Group 1: {selected_columns[0]}")
+        self.output_text.append(f"  Sample size (n₁): {len(data1)}")
+        self.output_text.append(f"  Median: {np.median(data1):.4f}")
+        self.output_text.append(f"  Mean rank: {stats.rankdata(np.concatenate([data1, data2]))[:len(data1)].mean():.2f}")
+        self.output_text.append("")
+        self.output_text.append(f"Group 2: {selected_columns[1]}")
+        self.output_text.append(f"  Sample size (n₂): {len(data2)}")
+        self.output_text.append(f"  Median: {np.median(data2):.4f}")
+        self.output_text.append(f"  Mean rank: {stats.rankdata(np.concatenate([data1, data2]))[-len(data2):].mean():.2f}")
+        self.output_text.append("")
+        self.output_text.append("TEST STATISTICS:")
+        self.output_text.append(f"  U-statistic: {statistic:.6f}")
+        self.output_text.append(f"  p-value: {p_value:.6f}")
+        self.output_text.append("")
+        
+        # Interpretation
+        alpha = 0.05
+        if p_value < alpha:
+            self.output_text.append(f"RESULT: Significant difference (p < {alpha})")
+        else:
+            self.output_text.append(f"RESULT: No significant difference (p ≥ {alpha})")
+            
+    def execute_correlation(self, selected_columns):
+        """Execute Pearson correlation analysis"""
+        if len(selected_columns) != 2:
+            self.show_error(f"Correlation analysis requires exactly 2 columns. You have selected {len(selected_columns)} columns.")
+            return
+            
+        # Extract numeric data from columns
+        data1 = self.get_numeric_column_data(selected_columns[0])
+        data2 = self.get_numeric_column_data(selected_columns[1])
+        
+        if data1 is None or len(data1) == 0:
+            self.show_error(f"Column '{selected_columns[0]}' contains no numeric data.")
+            return
+        if data2 is None or len(data2) == 0:
+            self.show_error(f"Column '{selected_columns[1]}' contains no numeric data.")
+            return
+            
+        # For correlation, we need paired data
+        if len(data1) != len(data2):
+            self.show_error("Correlation analysis requires equal sample sizes (paired observations).")
+            return
+            
+        # Perform Pearson correlation
+        correlation, p_value = stats.pearsonr(data1, data2)
+        
+        # Display results
+        self.output_text.clear()
+        self.output_text.append("=" * 50)
+        self.output_text.append("PEARSON CORRELATION ANALYSIS RESULTS")
+        self.output_text.append("=" * 50)
+        self.output_text.append(f"Variable 1: {selected_columns[0]}")
+        self.output_text.append(f"Variable 2: {selected_columns[1]}")
+        self.output_text.append(f"Sample size (n): {len(data1)}")
+        self.output_text.append("")
+        self.output_text.append("CORRELATION STATISTICS:")
+        self.output_text.append(f"  Pearson correlation coefficient (r): {correlation:.6f}")
+        self.output_text.append(f"  p-value: {p_value:.6f}")
+        self.output_text.append(f"  R-squared (r²): {correlation**2:.6f}")
+        self.output_text.append("")
+        
+        # Interpretation of correlation strength
+        abs_corr = abs(correlation)
+        if abs_corr < 0.1:
+            strength = "negligible"
+        elif abs_corr < 0.3:
+            strength = "weak"
+        elif abs_corr < 0.5:
+            strength = "moderate"
+        elif abs_corr < 0.7:
+            strength = "strong"
+        else:
+            strength = "very strong"
+            
+        direction = "positive" if correlation > 0 else "negative"
+        
+        self.output_text.append(f"INTERPRETATION:")
+        self.output_text.append(f"  Correlation strength: {strength}")
+        self.output_text.append(f"  Correlation direction: {direction}")
+        self.output_text.append("")
+        
+        # Statistical significance
+        alpha = 0.05
+        if p_value < alpha:
+            self.output_text.append(f"RESULT: Significant correlation (p < {alpha})")
+        else:
+            self.output_text.append(f"RESULT: No significant correlation (p ≥ {alpha})")
+            
+    def execute_normality_test(self, selected_columns):
+        """Execute Shapiro-Wilk normality test"""
+        if len(selected_columns) == 0:
+            self.show_error("Normality test requires at least 1 column.")
+            return
+            
+        self.output_text.clear()
+        self.output_text.append("=" * 50)
+        self.output_text.append("SHAPIRO-WILK NORMALITY TEST RESULTS")
+        self.output_text.append("=" * 50)
+        
+        for col_name in selected_columns:
+            data = self.get_numeric_column_data(col_name)
+            
+            if data is None or len(data) == 0:
+                self.output_text.append(f"Column '{col_name}': No numeric data available")
+                self.output_text.append("")
+                continue
+                
+            if len(data) < 3:
+                self.output_text.append(f"Column '{col_name}': Insufficient data (n < 3)")
+                self.output_text.append("")
+                continue
+                
+            if len(data) > 5000:
+                self.output_text.append(f"Column '{col_name}': Sample too large for Shapiro-Wilk (n > 5000)")
+                self.output_text.append("Consider using Kolmogorov-Smirnov test for large samples.")
+                self.output_text.append("")
+                continue
+                
+            # Perform Shapiro-Wilk test
+            statistic, p_value = stats.shapiro(data)
+            
+            self.output_text.append(f"Column: {col_name}")
+            self.output_text.append(f"  Sample size (n): {len(data)}")
+            self.output_text.append(f"  Mean: {np.mean(data):.4f}")
+            self.output_text.append(f"  Std Dev: {np.std(data, ddof=1):.4f}")
+            self.output_text.append(f"  Shapiro-Wilk statistic: {statistic:.6f}")
+            self.output_text.append(f"  p-value: {p_value:.6f}")
+            
+            # Interpretation
+            alpha = 0.05
+            if p_value < alpha:
+                self.output_text.append(f"  RESULT: Data significantly deviates from normal distribution (p < {alpha})")
+            else:
+                self.output_text.append(f"  RESULT: Data appears normally distributed (p ≥ {alpha})")
+            self.output_text.append("")
+            
+    def execute_chisquare(self, selected_columns):
+        """Execute Chi-square test of independence"""
+        if len(selected_columns) != 2:
+            self.show_error(f"Chi-square test requires exactly 2 columns. You have selected {len(selected_columns)} columns.")
+            return
+            
+        # Extract categorical data from columns
+        data1 = self.get_categorical_column_data(selected_columns[0])
+        data2 = self.get_categorical_column_data(selected_columns[1])
+        
+        if data1 is None or len(data1) == 0:
+            self.show_error(f"Column '{selected_columns[0]}' contains no data.")
+            return
+        if data2 is None or len(data2) == 0:
+            self.show_error(f"Column '{selected_columns[1]}' contains no data.")
+            return
+            
+        if len(data1) != len(data2):
+            self.show_error("Chi-square test requires equal sample sizes (paired observations).")
+            return
+            
+        # Create contingency table
+        try:
+            contingency_table = pd.crosstab(data1, data2)
+            
+            # Perform chi-square test
+            chi2, p_value, dof, expected = stats.chi2_contingency(contingency_table)
+            
+            # Display results
+            self.output_text.clear()
+            self.output_text.append("=" * 50)
+            self.output_text.append("CHI-SQUARE TEST OF INDEPENDENCE RESULTS")
+            self.output_text.append("=" * 50)
+            self.output_text.append(f"Variable 1: {selected_columns[0]}")
+            self.output_text.append(f"Variable 2: {selected_columns[1]}")
+            self.output_text.append(f"Sample size (n): {len(data1)}")
+            self.output_text.append("")
+            
+            self.output_text.append("CONTINGENCY TABLE:")
+            self.output_text.append(str(contingency_table))
+            self.output_text.append("")
+            
+            self.output_text.append("TEST STATISTICS:")
+            self.output_text.append(f"  Chi-square statistic: {chi2:.6f}")
+            self.output_text.append(f"  p-value: {p_value:.6f}")
+            self.output_text.append(f"  Degrees of freedom: {dof}")
+            self.output_text.append("")
+            
+            # Check assumptions
+            min_expected = np.min(expected)
+            cells_below_5 = np.sum(expected < 5)
+            total_cells = expected.size
+            
+            self.output_text.append("ASSUMPTION CHECK:")
+            self.output_text.append(f"  Minimum expected frequency: {min_expected:.2f}")
+            self.output_text.append(f"  Cells with expected frequency < 5: {cells_below_5}/{total_cells}")
+            
+            if min_expected < 1 or (cells_below_5 / total_cells) > 0.2:
+                self.output_text.append("  WARNING: Chi-square assumptions may be violated!")
+                self.output_text.append("  Consider Fisher's exact test for small samples.")
+            else:
+                self.output_text.append("  Chi-square assumptions satisfied.")
+            self.output_text.append("")
+            
+            # Calculate effect size (Cramér's V)
+            n = len(data1)
+            cramers_v = np.sqrt(chi2 / (n * (min(contingency_table.shape) - 1)))
+            self.output_text.append(f"EFFECT SIZE:")
+            self.output_text.append(f"  Cramér's V: {cramers_v:.4f}")
+            self.output_text.append("")
+            
+            # Interpretation
+            alpha = 0.05
+            if p_value < alpha:
+                self.output_text.append(f"RESULT: Significant association between variables (p < {alpha})")
+            else:
+                self.output_text.append(f"RESULT: No significant association between variables (p ≥ {alpha})")
+                
+        except Exception as e:
+            self.show_error(f"Error creating contingency table: {str(e)}")
+            
+    def show_error(self, message):
+        """Show error message"""
+        msg_box = QMessageBox()
+        msg_box.setIcon(QMessageBox.Icon.Critical)
+        msg_box.setWindowTitle("Error")
+        msg_box.setText(message)
+        msg_box.exec()
+        self.output_text.append(f"ERROR: {message}")
+
+def main(app=None):
+    if app is None:
+        app = QApplication(sys.argv)
+        should_exec = True
+    else:
+        should_exec = False
+    
+    window = StatisticalTestGUI()
+    window.show()
+    
+    if should_exec:
+        sys.exit(app.exec())
+    else:
+        return window
+
+if __name__ == "__main__":
+    main()