statslibx 0.1.1__py3-none-any.whl → 0.1.3__py3-none-any.whl

statslibx/inferential.py

@@ -1,7 +1,9 @@
+from dataclasses import dataclass
 import numpy as np
 import pandas as pd
-from typing import Optional, Union, Literal, List
+from typing import Optional, Union, Literal, List, Dict, Any
 from datetime import datetime
+from scipy import stats
 
 class InferentialStats:
     """
@@ -447,6 +449,183 @@ class InferentialStats:
             else:
                 interpretations.append(f"Se RECHAZA normalidad al {sig_level}% de significancia")
         return interpretations
+
+    def hypothesis_test(
+            self,
+            method: Literal["mean", "difference_mean", "proportion", "variance"] = "mean",
+            column1: str = None,
+            column2: str = None,
+            alpha: float = 0.05,
+            homoscedasticity: Literal["levene", "bartlett", "var_test"] = "levene"
+        ) -> Dict[str, Any]:
+
+            data = self.data
+
+            if column1 is None:
+                raise ValueError("Debes especificar 'column1'.")
+
+            x = data[column1].dropna()
+
+            if method in ["difference_mean", "variance"] and column2 is None:
+                raise ValueError("Para este método debes pasar 'column2'.")
+
+            y = data[column2].dropna() if column2 else None
+
+            # --- homoscedasticity test ---
+            homo_result = None
+            if method in ["difference_mean", "variance"]:
+                homo_result = self._homoscedasticity_test(x, y, homoscedasticity)
+
+            # --- MAIN HYPOTHESIS TESTS ---
+            if method == "mean":
+                # One-sample t-test
+                t_stat, p_value = stats.ttest_1samp(x, popmean=np.mean(x))
+                test_name = "One-sample t-test"
+
+            elif method == "difference_mean":
+                # Two-sample t-test
+                equal_var = homo_result["equal_var"]
+                t_stat, p_value = stats.ttest_ind(x, y, equal_var=equal_var)
+                test_name = "Two-sample t-test"
+
+            elif method == "proportion":
+                # Proportion test (z-test)
+                p_hat = np.mean(x)
+                n = len(x)
+                z_stat = (p_hat - 0.5) / np.sqrt(0.5 * 0.5 / n)
+                p_value = 2 * (1 - stats.norm.cdf(abs(z_stat)))
+                t_stat = z_stat
+                test_name = "Proportion Z-test"
+
+            elif method == "variance":
+                # Classic F-test
+                var_x = np.var(x, ddof=1)
+                var_y = np.var(y, ddof=1)
+                F = var_x / var_y
+                dfn = len(x) - 1
+                dfd = len(y) - 1
+
+                p_value = 2 * min(stats.f.cdf(F, dfn, dfd), 1 - stats.f.cdf(F, dfn, dfd))
+                t_stat = F
+                test_name = "Variance F-test"
+
+            return {
+                "test": test_name,
+                "statistic": t_stat,
+                "p_value": p_value,
+                "alpha": alpha,
+                "reject_H0": p_value < alpha,
+                "homoscedasticity_test": homo_result
+            }
+
+    def _homoscedasticity_test(
+        self,
+        x,
+        y,
+        method: Literal["levene", "bartlett", "var_test"] = "levene"
+    ) -> Dict[str, Any]:
+
+        if method == "levene":
+            stat, p = stats.levene(x, y)
+        elif method == "bartlett":
+            stat, p = stats.bartlett(x, y)
+        elif method == "var_test":
+            # R's var.test equivalent: F-test
+            var_x = np.var(x, ddof=1)
+            var_y = np.var(y, ddof=1)
+            F = var_x / var_y
+            dfn = len(x) - 1
+            dfd = len(y) - 1
+            p = 2 * min(stats.f.cdf(F, dfn, dfd), 1 - stats.f.cdf(F, dfn, dfd))
+            stat = F
+        else:
+            raise ValueError("Método de homocedasticidad no válido.")
+
+        return {
+            "method": method,
+            "statistic": stat,
+            "p_value": p,
+            "equal_var": p > 0.05   # estándar
+        }
+    
+    def variance_test(self, column1: str, column2: str,
+                    method: Literal['levene', 'bartlett', 'var_test'] = 'levene',
+                    center: Literal['mean', 'median', 'trimmed'] = 'median'
+                    ) -> 'TestResult':
+        """
+        Prueba de igualdad de varianzas entre dos columnas.
+
+        Parameters:
+        -----------
+        column1, column2 : str
+            Columnas numéricas a comparar
+        method : str
+            'levene'   -> robusto, recomendado cuando no se asume normalidad
+            'bartlett' -> muy sensible a normalidad
+            'var_test' -> equivalente a var.test de R (F-test)
+        center : str
+            Método de centrado para Levene ('mean', 'median', 'trimmed')
+
+        Returns:
+        --------
+        TestResult
+        """
+        from scipy import stats
+
+        data1 = self.data[column1].dropna().values
+        data2 = self.data[column2].dropna().values
+
+        if method == 'levene':
+            statistic, pvalue = stats.levene(data1, data2, center=center)
+            test_name = f'Test de Levene (center={center})'
+            params = {
+                'var1': data1.var(ddof=1),
+                'var2': data2.var(ddof=1),
+                'n1': len(data1), 'n2': len(data2)
+            }
+
+        elif method == 'bartlett':
+            statistic, pvalue = stats.bartlett(data1, data2)
+            test_name = 'Test de Bartlett'
+            params = {
+                'var1': data1.var(ddof=1),
+                'var2': data2.var(ddof=1),
+                'n1': len(data1), 'n2': len(data2)
+            }
+
+        elif method == 'var_test':
+            # F-test clásico de comparación de varianzas
+            var1 = data1.var(ddof=1)
+            var2 = data2.var(ddof=1)
+            f_stat = var1 / var2
+            df1 = len(data1) - 1
+            df2 = len(data2) - 1
+
+            # p-valor bilateral
+            pvalue = 2 * min(
+                stats.f.cdf(f_stat, df1, df2),
+                1 - stats.f.cdf(f_stat, df1, df2)
+            )
+
+            statistic = f_stat
+            test_name = 'F-test de Varianzas (var.test estilo R)'
+            params = {
+                'var1': var1, 'var2': var2,
+                'ratio': f_stat,
+                'df1': df1, 'df2': df2
+            }
+
+        else:
+            raise ValueError(f"Método '{method}' no válido. Usa levene, bartlett o var_test.")
+
+        return TestResult(
+            test_name=test_name,
+            statistic=statistic,
+            pvalue=pvalue,
+            alternative='two-sided',
+            params=params
+        )
+
     
     def help(self):
         """
@@ -514,6 +693,12 @@ class InferentialStats:
   • .mann_whitney_test(column1, column2, alternative='two-sided')
     Alternativa no paramétrica al t-test de dos muestras
 
+  🔹 Pruebas Extras:
+  • .hypothesis_test(method='mean', column1=None, column2=None, 
+                   alpha=0.05, homoscedasticity='levene')
+  • .variance_test(column1, column2, method='levene', center='median')
+    
+
 ┌────────────────────────────────────────────────────────────────────────────┐
 │ 4. 🧪 PRUEBAS PARA MÚLTIPLES GRUPOS                                        │
 └────────────────────────────────────────────────────────────────────────────┘
@@ -737,7 +922,8 @@ class InferentialStats:
 ╚════════════════════════════════════════════════════════════════════════════╝
     """
         print(help_text)
-    
+
+@dataclass
 class TestResult:
     """Clase para resultados de pruebas de hipótesis"""
     

statslibx/utils.py

@@ -602,20 +602,18 @@ class UtilsStats:
 
     # ============= GRÁFICOS CON INTERVALOS DE CONFIANZA =============
 
-    def plot_distribution_with_ci(self, 
-                                 data: Union[pd.DataFrame, pd.Series, np.ndarray],
-                                 column: Optional[str] = None,
-                                 confidence_level: float = 0.95,
-                                 ci_method: str = 'parametric',
-                                 bins: int = 30,
-                                 figsize: Optional[Tuple[int, int]] = None,
-                                 save_fig: Optional[bool] = None,
-                                 filename: Optional[str] = None,
-                                 **kwargs) -> plt.Figure:
-        """
-        Grafica la distribución junto con intervalos de confianza
-        """
-        # Extraer y limpiar datos
+    def plot_distribution_with_ci(self,
+                               data: Union[pd.DataFrame, pd.Series, np.ndarray],
+                               column: Optional[str] = None,
+                               confidence_level: float = 0.95,
+                               ci_method: str = 'parametric',
+                               bins: int = 30,
+                               figsize: Optional[Tuple[int, int]] = None,
+                               save_fig: Optional[bool] = None,
+                               filename: Optional[str] = None,
+                               **kwargs) -> plt.Figure:
+
+        # ======= PREPARACIÓN =======
         if isinstance(data, pd.DataFrame):
             if column is None:
                 raise ValueError("Debe especificar 'column' cuando data es DataFrame")
@@ -627,98 +625,91 @@ class UtilsStats:
         else:
             plot_data = pd.Series(data).dropna()
             data_name = 'Variable'
-        
+
         data_array = plot_data.values
-        default_filename = f"distribucion_ci_{data_name.lower().replace(' ', '_')}"
-        filename = filename or default_filename
-        
-        # Calcular estadísticas e intervalos de confianza
+        filename = filename or f"distribucion_ci_{data_name.lower().replace(' ', '_')}"
+
+        # Estadísticas
         ci_result = self.calculate_confidence_intervals(data_array, confidence_level, ci_method)
         normality_result = self.check_normality(data_array)
-        
-        # Crear figura con dos subgráficas
-        fig, (ax1, ax2) = plt.subplots(1, 2, figsize=figsize or (14, 6))
-        
-        # ===== PRIMERA GRÁFICA: Distribución básica =====
-        n, bins, patches = ax1.hist(data_array, bins=bins, alpha=0.7, 
-                                   color='skyblue', edgecolor='black', 
-                                   density=True, label='Histograma')
-        
+
         # KDE
         kde = stats.gaussian_kde(data_array)
-        x_range = np.linspace(data_array.min(), data_array.max(), 200)
+        x_range = np.linspace(data_array.min(), data_array.max(), 300)
+
+        # ======= FIGURA =======
+        fig, (ax1, ax2) = plt.subplots(1, 2, figsize=figsize or (14, 6))
+
+        # ============================================================
+        # PANEL 1: HISTOGRAMA + KDE
+        # ============================================================
+        ax1.hist(data_array, bins=bins, density=True,
+                color='skyblue', edgecolor='black', alpha=0.7)
+
         ax1.plot(x_range, kde(x_range), 'r-', linewidth=2, label='KDE')
-        
-        # Línea vertical en la media
-        ax1.axvline(ci_result['mean'], color='red', linestyle='--', 
-                   linewidth=2, label=f'Media: {ci_result["mean"]:.2f}')
-        
-        ax1.set_xlabel('Valores')
-        ax1.set_ylabel('Densidad')
-        ax1.set_title(f'Distribución de {data_name}\n'
-                     f'Media: {ci_result["mean"]:.2f}, '
-                     f'Desv. Est.: {ci_result["std"]:.2f}')
+
+        ax1.axvline(ci_result['mean'], color='red', linestyle='--', linewidth=2,
+                    label=f"Media: {ci_result['mean']:.2f}")
+
+        ax1.set_title(f"Distribución de {data_name}")
+        ax1.set_xlabel("Valores")
+        ax1.set_ylabel("Densidad")
         ax1.legend()
-        ax1.grid(True, alpha=0.3)
-        
-        # ===== SEGUNDA GRÁFICA: Distribución con intervalos de confianza =====
-        n, bins, patches = ax2.hist(data_array, bins=bins, alpha=0.7, 
-                                   color='lightgreen', edgecolor='black', 
-                                   density=True, label='Histograma')
-        
-        # KDE
+        ax1.grid(alpha=0.3)
+
+        # ============================================================
+        # PANEL 2: KDE + INTERVALO DE CONFIANZA
+        # ============================================================
+
+        # KDE pura
         ax2.plot(x_range, kde(x_range), 'r-', linewidth=2, label='KDE')
-        
-        # Media y intervalos de confianza
-        ax2.axvline(ci_result['mean'], color='red', linestyle='-', 
-                   linewidth=3, label=f'Media: {ci_result["mean"]:.2f}')
-        
-        # Intervalo de confianza
-        ax2.axvspan(ci_result['ci_lower'], ci_result['ci_upper'], 
-                   alpha=0.3, color='orange', 
-                   label=f'IC {confidence_level*100}%: [{ci_result["ci_lower"]:.2f}, {ci_result["ci_upper"]:.2f}]')
-        
-        # Líneas para los límites del IC
-        ax2.axvline(ci_result['ci_lower'], color='orange', linestyle='--', linewidth=2)
-        ax2.axvline(ci_result['ci_upper'], color='orange', linestyle='--', linewidth=2)
-        
-        # Distribución normal teórica (si los datos son normales)
-        if normality_result['is_normal']:
-            normal_x = np.linspace(data_array.min(), data_array.max(), 200)
-            normal_y = stats.norm.pdf(normal_x, ci_result['mean'], ci_result['std'])
-            ax2.plot(normal_x, normal_y, 'g--', linewidth=2, alpha=0.7, 
-                    label='Distribución Normal Teórica')
-        
-        ax2.set_xlabel('Valores')
-        ax2.set_ylabel('Densidad')
-        ax2.set_title(f'Distribución con Intervalos de Confianza\n'
-                     f'Método: {ci_method}, n={ci_result["n"]}')
+
+        # Intervalo de Confianza
+        ax2.axvspan(ci_result["ci_lower"], ci_result["ci_upper"],
+                    color='orange', alpha=0.3,
+                    label=f"IC {confidence_level*100:.0f}%")
+
+        # Media
+        ax2.axvline(ci_result["mean"], color='red', linewidth=2)
+
+        # Distribución normal teórica (si aplica)
+        if normality_result["is_normal"]:
+            normal_y = stats.norm.pdf(x_range, ci_result['mean'], ci_result['std'])
+            ax2.plot(x_range, normal_y, 'g--', linewidth=2, alpha=0.7,
+                    label="Normal Teórica")
+
+        ax2.set_title(f"IC con método '{ci_method}'")
+        ax2.set_xlabel("Valores")
+        ax2.set_ylabel("Densidad")
         ax2.legend()
-        ax2.grid(True, alpha=0.3)
-        
-        # Información adicional como texto
-        info_text = (f'Estadísticas:\n'
-                    f'• Media: {ci_result["mean"]:.3f}\n'
-                    f'• Desv. Est.: {ci_result["std"]:.3f}\n'
-                    f'• n: {ci_result["n"]}\n'
-                    f'• IC {confidence_level*100}%: [{ci_result["ci_lower"]:.3f}, {ci_result["ci_upper"]:.3f}]\n'
-                    f'• Margen Error: ±{ci_result["margin_error"]:.3f}\n'
-                    f'• Normalidad: {normality_result["interpretation"]}\n'
-                    f'• p-value: {normality_result["shapiro_pvalue"]:.4f}')
-        
-        fig.text(0.02, 0.02, info_text, fontsize=9, 
-                bbox=dict(boxstyle="round,pad=0.5", facecolor="lightgray", alpha=0.7),
-                verticalalignment='bottom')
-        
+        ax2.grid(alpha=0.3)
+
+        # ======= CUADRO DE INFO =======
+        info = (
+            f"Estadísticas de {data_name}:\n"
+            f"• n = {ci_result['n']}\n"
+            f"• Media = {ci_result['mean']:.3f}\n"
+            f"• Desv. Est. = {ci_result['std']:.3f}\n"
+            f"• IC {confidence_level*100:.0f}% = [{ci_result['ci_lower']:.3f}, {ci_result['ci_upper']:.3f}]\n"
+            f"• Margen Error = ±{ci_result['margin_error']:.3f}\n"
+            f"• Normalidad = {normality_result['interpretation']}\n"
+            f"• p-value Shapiro = {normality_result['shapiro_pvalue']:.4f}"
+        )
+
+        fig.text(0.01, 0.01, info, fontsize=9,
+                bbox=dict(facecolor='lightgray', alpha=0.6),
+                va='bottom')
+
         plt.tight_layout()
-        
-        # Guardar figura si está activado
+
+        # Guardado opcional
         save_fig = save_fig if save_fig is not None else self._save_fig
         if save_fig:
             self._save_figure(fig, filename)
-        
+
         return fig
 
+
     def plot_multiple_distributions_with_ci(self, 
                                            data_dict: dict,
                                            confidence_level: float = 0.95,
@@ -877,7 +868,8 @@ class UtilsStats:
     backend: 'matplotlib', 'seaborn', 'plotly'
 
   • .plot_distribution_with_ci(data, column=None, confidence_level=0.95,
-                               ci_method='parametric', bins=30)
+                               ci_method='parametric', bins=30, figsize=None,
+                               save_fig=None, filename=None)
     
     Distribución con intervalos de confianza visualizados
 

{statslibx-0.1.1.dist-info → statslibx-0.1.3.dist-info}/METADATA

@@ -1,8 +1,8 @@
 Metadata-Version: 2.4
 Name: statslibx
-Version: 0.1.1
+Version: 0.1.3
 Summary: Librería de estadística descriptiva e inferencial para Python
-Home-page: https://github.com/Immanuel3008/StatsLibrary
+Home-page: https://github.com/Immanuel3008/StatsLibX
 Author: Emmanuel Ascendra Perez
 Author-email: ascendraemmanuel@gmail.com
 Classifier: Development Status :: 3 - Alpha
@@ -43,4 +43,39 @@ Dynamic: requires-dist
 Dynamic: requires-python
 Dynamic: summary
 
-Librería de estadística descriptiva e inferencial para Python
+📦 Descripción para PyPI (Plantilla Profesional)
+
+StatsLibX es un paquete de Python diseñado para proporcionar una solución sencilla, eficiente y flexible para manejar volumenes de datos.
+
+Este proyecto surge con la idea de ofrecer una alternativa moderna, intuitiva y ligera que permita a desarrolladores y entusiastas integrar la estadistica descriptiva e inferencial sin complicaciones, con multiples funcionalidades y utilidades pensadas para el futuro.
+
+✨ Características principales
+
+⚡ Rápido y eficiente: optimizado para ofrecer un rendimiento suave incluso en tareas exigentes.
+
+🧩 Fácil de usar: una API limpia para que empieces en segundos.
+
+🔧 Altamente extensible: personalízalo según tus necesidades.
+
+📚 Documentación clara: ejemplos simples y prácticos.
+
+🔮 Diseñado con visión a futuro: construido para escalar y adaptarse.
+
+🚀 Ejemplo rápido
+```python
+from statslibx import DescriptiveStats, InferentialStats, UtilsStats
+
+stats = DescriptiveStats(data) # InferentialStats(data), UtilsStats()
+stats.help()
+```
+
+📦 Instalación
+```bash
+pip install statslibx
+```
+
+🤝 Contribuciones
+
+¡Todas las mejoras e ideas son bienvenidas! 
+
+E-mail: ascendraemmanuel@gmail.com

statslibx-0.1.3.dist-info/RECORD

@@ -0,0 +1,12 @@
+statslibx/__init__.py,sha256=FRvBhSxEoGyg548eOe23UqKpabWQwVv0LylMm2oBdek,1173
+statslibx/descriptive.py,sha256=Hu7VuOGXs6oOq-zxQNiBKg7UtkNdNQ1Qy3PP-wEO5_k,36971
+statslibx/inferential.py,sha256=BVBxEdLnNCw2yC-3s5fZ84oeJ8LqJYR_IJquPEiyiOk,48234
+statslibx/utils.py,sha256=tdf1yZuR4fsmNq24ygv69BgCLzB0iE_x0ki1IV7Iwxs,60693
+statslibx/datasets/__init__.py,sha256=DtXI31_S7A-D0KFDo3b_7_f6rTMQLEHtzEQRQxs12Xw,337
+statslibx/datasets/iris.csv,sha256=xSdC5QMVqZ-Vajg_rt91dVUmdfZAnvD5pHB23QhHmTA,3858
+statslibx/datasets/penguins.csv,sha256=4HY2vYr3QmAJnqL4Z44uq7813vV5lAzHb2cGHuFsBsE,13478
+statslibx/datasets/sp500_companies.csv,sha256=WKS72YOGnAbyLR6kD95fOpIYZt5oXGjPryyFVqLRF_k,803820
+statslibx-0.1.3.dist-info/METADATA,sha256=p4y_4wrDCL6agGm1-EID1EL3JbglQgUoC15jtQyo7E8,2917
+statslibx-0.1.3.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+statslibx-0.1.3.dist-info/top_level.txt,sha256=eeYZXyFm0hIjuI0ba3wF6XW938Mv9tv7Nk9qgjYfCtU,10
+statslibx-0.1.3.dist-info/RECORD,,

statslibx-0.1.1.dist-info/RECORD

@@ -1,8 +0,0 @@
-statslibx/__init__.py,sha256=p1AydN8u_zWgm1rJBJb8TVQkRUzQMA3iNDwfFZn8k00,1117
-statslibx/descriptive.py,sha256=Hu7VuOGXs6oOq-zxQNiBKg7UtkNdNQ1Qy3PP-wEO5_k,36971
-statslibx/inferential.py,sha256=slLh32Ny4doLA0EA8pYRUGQSuMI8oBUCMBu-CTX-7FY,41732
-statslibx/utils.py,sha256=vnfs5LmWEKsB9p8Fs2Di3btReepkB1RYAwQfT-eZs6c,61856
-statslibx-0.1.1.dist-info/METADATA,sha256=ujX3UiJWx5ibgNv7OaJevK5YYeIaxudPkKzlFWBChO0,1737
-statslibx-0.1.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-statslibx-0.1.1.dist-info/top_level.txt,sha256=eeYZXyFm0hIjuI0ba3wF6XW938Mv9tv7Nk9qgjYfCtU,10
-statslibx-0.1.1.dist-info/RECORD,,