{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "###

Complementaria 4: Manejo de datos y análisis preliminar en Python

" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "El objetivo de esta complementaria es aprender a procesar datos en Python, desde importar y exportar archivos, hasta el manejo interno de datos y el ajuste de distribuciones." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "

Importar archivos de datos\n", "

" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Para importar archivos de datos de extensión csv, txt, xlsx, dta o sas, podemos utilizar la librería `pandas`." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "A continuación, queremos importar la base de datos “AAPL.csv” que se encuentra en Bloque Neón. Para esto, usaremos la función `pd.read_csv()`. Antes de ejecutar el código para importar el archivo, debe establecer el directorio de trabajo (Working Directory) en la ruta en la que se encuentra el archivo en su computador. (P. ej. “C://Desktop/Modelos 202510/Complementarias/Complementaria 4/AAPL.csv”) utilizando el módulo `os`." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DateOpenHighLowCloseAdj CloseVolume
02017-07-17148.820007150.899994148.570007149.559998148.97497623793500
12017-07-18149.199997150.130005148.669998150.080002149.49295017868800
22017-07-19150.479996151.419998149.949997151.020004150.42927620923000
32017-07-20151.500000151.740005150.190002150.339996149.75192317243700
42017-07-21149.990005150.440002148.880005150.270004149.68220526252600
\n", "
" ], "text/plain": [ " Date Open High Low Close Adj Close \\\n", "0 2017-07-17 148.820007 150.899994 148.570007 149.559998 148.974976 \n", "1 2017-07-18 149.199997 150.130005 148.669998 150.080002 149.492950 \n", "2 2017-07-19 150.479996 151.419998 149.949997 151.020004 150.429276 \n", "3 2017-07-20 151.500000 151.740005 150.190002 150.339996 149.751923 \n", "4 2017-07-21 149.990005 150.440002 148.880005 150.270004 149.682205 \n", "\n", " Volume \n", "0 23793500 \n", "1 17868800 \n", "2 20923000 \n", "3 17243700 \n", "4 26252600 " ] }, "execution_count": 1, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import os\n", "import pandas as pd\n", "\n", "#Conocer el directorio de trabajo actual\n", "directorio_actual = os.getcwd()\n", "#print(directorio_actual)\n", "\n", "#Si se quiere cambiar el directorio\n", "'''os.chdir('/ruta/del/nuevo/directorio')'''\n", "\n", "df = pd.read_csv('AAPL.csv')\n", "df.head()\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "La base de datos “AAPL.csv” incluye la información de las acciones de Apple para los meses de julio y agosto del año 2017. Para cada fecha se conoce el precio al que abrió la acción ese día, los valores máximo y mínimo del precio, el precio de cierre de la acción, el precio ajustado y, finalmente, el volumen de acciones transado. De esta misma manera se pueden importar datos de archivos con diferente extensión (`pd.read_excel()`, `pd.read_stata()`, por ejemplo)." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "*Nota:* Para conocer el directorio actual dónde se está trabajando en Python se puede utilizar la función `getcwd()` del módulo `os`. Si se desea cambiar el directorio de trabajo se puede usar la función chdir() ingresando como parámetro la dirección que se quiere asignar, por ejemplo:\n", "`os.chdir(\"C:Desktop/Modelos 202510/Complementarias/Complementaria 4\").`" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "

DataFrame\n", "

" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Cuando importamos los datos utilizando la librería de `pandas` estos se almacenan en estructura de DataFrame. Un DataFrame es útil porque podemos almacenar vectores de diferente tipo (por ejemplo, numéricos y caracteres). También es útil porque podemos acceder a la información de cierta columna, conociendo el número de la columna o el nombre que le asignamos a dicha columna. La línea superior del DataFrame, llamada el encabezado, contiene los nombres de columna. Cada línea horizontal denota una fila de datos, que comienza con el nombre de la fila y, a continuación, los datos." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Para llamar los datos de una celda se pueden utilizar dos métodos diferentes:\n", "* iloc[]: Utiliza índices numéricos (enteros) para acceder a las filas y las columnas.\n", "* loc[]: Utiliza etiquetas de fila y nombres de columnas para acceder a los datos." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Para llamar los datos de una celda utilizando `iloc[]` introducimos sus coordenadas de fila y columna (como números enteros) en el operador de corchetes [ ]. Las dos coordenadas están separadas por una coma [numero de fila, numero de columna]. Por ejemplo, queremos traer el dato que se encuentra en la fila 3 y columna 4 (recuerde que los índices en Python empiezan en 0, por lo que en este caso se debe usar el índice 2 para las filas y el 3 para las columnas)." ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "np.float64(149.949997)" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.iloc[2,3]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Para llamar los datos de una celda utilizando `loc[]` introducimos la etiqueta de la fila (si no tiene etiqueta es directamente el número de fila) y el nombre de la columna en el operador de corchetes [ ]." ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "np.float64(149.949997)" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.loc[2,'Low']" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "De igual forma, podemos conocer el número de filas y columnas de un DataFrame con las función `shape`. Esta función guarda las dimensiones del DataFrame en una tupla, en donde el primer elemento corresponde al número de filas y el segundo almacena el número de columnas." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(21, 7)" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.shape" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "El DataFrame tiene 21 filas.\n" ] } ], "source": [ "num_filas = df.shape[0]\n", "print(f'El DataFrame tiene {df.shape[0]} filas.')" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "El DataFrame tiene 7 columnas.\n" ] } ], "source": [ "num_columnas = df.shape[1]\n", "print(f'El DataFrame tiene {df.shape[1]} columnas.')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Más aún, podemos acceder a la información básica del DataFrame utilizando la función `describe()`:" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
OpenHighLowCloseAdj CloseVolume
count21.00000021.00000021.00000021.00000021.0000002.100000e+01
mean153.880954154.989048152.591431153.814287153.3006632.657169e+07
std4.0531364.2524133.9806174.0471534.1221801.205136e+07
min148.820007150.130005147.300003148.729996148.1482241.578100e+07
25%149.990005150.899994148.880005150.270004149.6822051.984590e+07
50%153.350006153.929993151.800003152.740005152.1425482.202820e+07
75%157.059998158.919998156.070007157.139999156.5253302.709730e+07
max159.899994161.830002159.110001161.059998160.4299936.993680e+07
\n", "
" ], "text/plain": [ " Open High Low Close Adj Close \\\n", "count 21.000000 21.000000 21.000000 21.000000 21.000000 \n", "mean 153.880954 154.989048 152.591431 153.814287 153.300663 \n", "std 4.053136 4.252413 3.980617 4.047153 4.122180 \n", "min 148.820007 150.130005 147.300003 148.729996 148.148224 \n", "25% 149.990005 150.899994 148.880005 150.270004 149.682205 \n", "50% 153.350006 153.929993 151.800003 152.740005 152.142548 \n", "75% 157.059998 158.919998 156.070007 157.139999 156.525330 \n", "max 159.899994 161.830002 159.110001 161.059998 160.429993 \n", "\n", " Volume \n", "count 2.100000e+01 \n", "mean 2.657169e+07 \n", "std 1.205136e+07 \n", "min 1.578100e+07 \n", "25% 1.984590e+07 \n", "50% 2.202820e+07 \n", "75% 2.709730e+07 \n", "max 6.993680e+07 " ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.describe()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Allí encontramos las estadísticas básicas de variables numéricas (media, cuartiles, desviación, mínimo y máximo) para cada columna dentro del DataFrame." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Para acceder a una columna podemos hacerlo de varias formas:\n", "1. Utilizando el nombre de la columna.\n", "2. Utilizando el atributo de columna.\n", "3. Utilizando la función `loc[]`.\n", "4. Utilizando la función `iloc[]`." ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0 23793500\n", "1 17868800\n", "2 20923000\n", "3 17243700\n", "4 26252600\n", "5 21493200\n", "6 18853900\n", "7 15781000\n", "8 32476300\n", "9 17213700\n", "10 19845900\n", "11 35368600\n", "12 69936800\n", "13 27097300\n", "14 20559900\n", "15 21870300\n", "16 36205900\n", "17 26131500\n", "18 40804300\n", "19 26257100\n", "20 22028200\n", "Name: Volume, dtype: int64" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df['Volume']" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0 23793500\n", "1 17868800\n", "2 20923000\n", "3 17243700\n", "4 26252600\n", "5 21493200\n", "6 18853900\n", "7 15781000\n", "8 32476300\n", "9 17213700\n", "10 19845900\n", "11 35368600\n", "12 69936800\n", "13 27097300\n", "14 20559900\n", "15 21870300\n", "16 36205900\n", "17 26131500\n", "18 40804300\n", "19 26257100\n", "20 22028200\n", "Name: Volume, dtype: int64" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Para este método el nombre de la columna no puede tener espacios ni caracteres especiales\n", "df.Volume" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0 23793500\n", "1 17868800\n", "2 20923000\n", "3 17243700\n", "4 26252600\n", "5 21493200\n", "6 18853900\n", "7 15781000\n", "8 32476300\n", "9 17213700\n", "10 19845900\n", "11 35368600\n", "12 69936800\n", "13 27097300\n", "14 20559900\n", "15 21870300\n", "16 36205900\n", "17 26131500\n", "18 40804300\n", "19 26257100\n", "20 22028200\n", "Name: Volume, dtype: int64" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.loc[:,'Volume']" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0 23793500\n", "1 17868800\n", "2 20923000\n", "3 17243700\n", "4 26252600\n", "5 21493200\n", "6 18853900\n", "7 15781000\n", "8 32476300\n", "9 17213700\n", "10 19845900\n", "11 35368600\n", "12 69936800\n", "13 27097300\n", "14 20559900\n", "15 21870300\n", "16 36205900\n", "17 26131500\n", "18 40804300\n", "19 26257100\n", "20 22028200\n", "Name: Volume, dtype: int64" ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.iloc[:,6]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Si se quiere acceder a múltiples columnas simultáneamente se puede hacer uso del doble corchete [[]]." ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
HighVolume
0150.89999423793500
1150.13000517868800
2151.41999820923000
3151.74000517243700
4150.44000226252600
5152.44000221493200
6153.83999618853900
7153.92999315781000
8153.99000532476300
9150.22999617213700
10150.33000219845900
11150.22000135368600
12159.75000069936800
13157.21000727097300
14157.39999420559900
15158.91999821870300
16161.83000236205900
17161.27000426131500
18160.00000040804300
19158.57000726257100
20160.21000722028200
\n", "
" ], "text/plain": [ " High Volume\n", "0 150.899994 23793500\n", "1 150.130005 17868800\n", "2 151.419998 20923000\n", "3 151.740005 17243700\n", "4 150.440002 26252600\n", "5 152.440002 21493200\n", "6 153.839996 18853900\n", "7 153.929993 15781000\n", "8 153.990005 32476300\n", "9 150.229996 17213700\n", "10 150.330002 19845900\n", "11 150.220001 35368600\n", "12 159.750000 69936800\n", "13 157.210007 27097300\n", "14 157.399994 20559900\n", "15 158.919998 21870300\n", "16 161.830002 36205900\n", "17 161.270004 26131500\n", "18 160.000000 40804300\n", "19 158.570007 26257100\n", "20 160.210007 22028200" ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[['High','Volume']]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "También, podemos acceder a la información de las filas de diferentes formas:" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Date 2017-07-19\n", "Open 150.479996\n", "High 151.419998\n", "Low 149.949997\n", "Close 151.020004\n", "Adj Close 150.429276\n", "Volume 20923000\n", "Name: 2, dtype: object" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.iloc[2,]" ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "\"df.loc['Nombre de la fila',]\"" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "'''df.loc['Nombre de la fila',]'''" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Podemos traer información de múltiples filas o columnas. En este caso queremos traer la información de las filas 5 y 12 al mismo tiempo:" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DateOpenHighLowCloseAdj CloseVolume
42017-07-21149.990005150.440002148.880005150.270004149.68220526252600
112017-08-01149.100006150.220001148.410004150.050003149.46305835368600
\n", "
" ], "text/plain": [ " Date Open High Low Close Adj Close \\\n", "4 2017-07-21 149.990005 150.440002 148.880005 150.270004 149.682205 \n", "11 2017-08-01 149.100006 150.220001 148.410004 150.050003 149.463058 \n", "\n", " Volume \n", "4 26252600 \n", "11 35368600 " ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.iloc[[4,11]]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Por último, es posible obtener un subconjunto de filas dentro del data frame que cumplan cierta condición haciendo uso de arreglos de operadores lógicos." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Cuando solo se tiene una condición:" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DateOpenHighLowCloseAdj CloseVolume
22017-07-19150.479996151.419998149.949997151.020004150.42927620923000
52017-07-24150.580002152.440002149.899994152.089996151.49507121493200
62017-07-25151.800003153.839996151.800003152.740005152.14254818853900
72017-07-26153.350006153.929993153.059998153.460007152.85972615781000
122017-08-02159.279999159.750000156.160004157.139999156.52533069936800
132017-08-03157.050003157.210007155.020004155.570007154.96147227097300
142017-08-04156.070007157.399994155.690002156.389999155.77825920559900
152017-08-07157.059998158.919998156.669998158.809998158.18879721870300
162017-08-08158.600006161.830002158.270004160.080002159.45382736205900
172017-08-09159.259995161.270004159.110001161.059998160.42999326131500
182017-08-10159.899994160.000000154.630005155.320007155.32000740804300
192017-08-11156.600006158.570007156.070007157.479996157.47999626257100
202017-08-14159.320007160.210007158.750000159.850006159.85000622028200
\n", "
" ], "text/plain": [ " Date Open High Low Close Adj Close \\\n", "2 2017-07-19 150.479996 151.419998 149.949997 151.020004 150.429276 \n", "5 2017-07-24 150.580002 152.440002 149.899994 152.089996 151.495071 \n", "6 2017-07-25 151.800003 153.839996 151.800003 152.740005 152.142548 \n", "7 2017-07-26 153.350006 153.929993 153.059998 153.460007 152.859726 \n", "12 2017-08-02 159.279999 159.750000 156.160004 157.139999 156.525330 \n", "13 2017-08-03 157.050003 157.210007 155.020004 155.570007 154.961472 \n", "14 2017-08-04 156.070007 157.399994 155.690002 156.389999 155.778259 \n", "15 2017-08-07 157.059998 158.919998 156.669998 158.809998 158.188797 \n", "16 2017-08-08 158.600006 161.830002 158.270004 160.080002 159.453827 \n", "17 2017-08-09 159.259995 161.270004 159.110001 161.059998 160.429993 \n", "18 2017-08-10 159.899994 160.000000 154.630005 155.320007 155.320007 \n", "19 2017-08-11 156.600006 158.570007 156.070007 157.479996 157.479996 \n", "20 2017-08-14 159.320007 160.210007 158.750000 159.850006 159.850006 \n", "\n", " Volume \n", "2 20923000 \n", "5 21493200 \n", "6 18853900 \n", "7 15781000 \n", "12 69936800 \n", "13 27097300 \n", "14 20559900 \n", "15 21870300 \n", "16 36205900 \n", "17 26131500 \n", "18 40804300 \n", "19 26257100 \n", "20 22028200 " ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[df['Adj Close']>150]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Cuando se tienen múltiples condiciones:" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DateOpenHighLowCloseAdj CloseVolume
22017-07-19150.479996151.419998149.949997151.020004150.42927620923000
52017-07-24150.580002152.440002149.899994152.089996151.49507121493200
\n", "
" ], "text/plain": [ " Date Open High Low Close Adj Close \\\n", "2 2017-07-19 150.479996 151.419998 149.949997 151.020004 150.429276 \n", "5 2017-07-24 150.580002 152.440002 149.899994 152.089996 151.495071 \n", "\n", " Volume \n", "2 20923000 \n", "5 21493200 " ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[(df['Adj Close']>150) & (df['Low']<150)]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Ahora bien, también podemos añadir columnas a un DataFrame. Por ejemplo, podemos crear una columna, `LogAdj`, que sea el logaritmo del precio ajustado de cierre:" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DateOpenHighLowCloseAdj CloseVolumeLogAdj
02017-07-17148.820007150.899994148.570007149.559998148.974976237935002.173113
12017-07-18149.199997150.130005148.669998150.080002149.492950178688002.174621
22017-07-19150.479996151.419998149.949997151.020004150.429276209230002.177332
32017-07-20151.500000151.740005150.190002150.339996149.751923172437002.175372
42017-07-21149.990005150.440002148.880005150.270004149.682205262526002.175170
52017-07-24150.580002152.440002149.899994152.089996151.495071214932002.180399
62017-07-25151.800003153.839996151.800003152.740005152.142548188539002.182251
72017-07-26153.350006153.929993153.059998153.460007152.859726157810002.184293
82017-07-27153.750000153.990005147.300003150.559998149.971069324763002.176007
92017-07-28149.889999150.229996149.190002149.500000148.915207172137002.172939
102017-07-31149.899994150.330002148.130005148.729996148.148224198459002.170696
112017-08-01149.100006150.220001148.410004150.050003149.463058353686002.174534
122017-08-02159.279999159.750000156.160004157.139999156.525330699368002.194585
132017-08-03157.050003157.210007155.020004155.570007154.961472270973002.190224
142017-08-04156.070007157.399994155.690002156.389999155.778259205599002.192507
152017-08-07157.059998158.919998156.669998158.809998158.188797218703002.199176
162017-08-08158.600006161.830002158.270004160.080002159.453827362059002.202635
172017-08-09159.259995161.270004159.110001161.059998160.429993261315002.205286
182017-08-10159.899994160.000000154.630005155.320007155.320007408043002.191227
192017-08-11156.600006158.570007156.070007157.479996157.479996262571002.197225
202017-08-14159.320007160.210007158.750000159.850006159.850006220282002.203713
\n", "
" ], "text/plain": [ " Date Open High Low Close Adj Close \\\n", "0 2017-07-17 148.820007 150.899994 148.570007 149.559998 148.974976 \n", "1 2017-07-18 149.199997 150.130005 148.669998 150.080002 149.492950 \n", "2 2017-07-19 150.479996 151.419998 149.949997 151.020004 150.429276 \n", "3 2017-07-20 151.500000 151.740005 150.190002 150.339996 149.751923 \n", "4 2017-07-21 149.990005 150.440002 148.880005 150.270004 149.682205 \n", "5 2017-07-24 150.580002 152.440002 149.899994 152.089996 151.495071 \n", "6 2017-07-25 151.800003 153.839996 151.800003 152.740005 152.142548 \n", "7 2017-07-26 153.350006 153.929993 153.059998 153.460007 152.859726 \n", "8 2017-07-27 153.750000 153.990005 147.300003 150.559998 149.971069 \n", "9 2017-07-28 149.889999 150.229996 149.190002 149.500000 148.915207 \n", "10 2017-07-31 149.899994 150.330002 148.130005 148.729996 148.148224 \n", "11 2017-08-01 149.100006 150.220001 148.410004 150.050003 149.463058 \n", "12 2017-08-02 159.279999 159.750000 156.160004 157.139999 156.525330 \n", "13 2017-08-03 157.050003 157.210007 155.020004 155.570007 154.961472 \n", "14 2017-08-04 156.070007 157.399994 155.690002 156.389999 155.778259 \n", "15 2017-08-07 157.059998 158.919998 156.669998 158.809998 158.188797 \n", "16 2017-08-08 158.600006 161.830002 158.270004 160.080002 159.453827 \n", "17 2017-08-09 159.259995 161.270004 159.110001 161.059998 160.429993 \n", "18 2017-08-10 159.899994 160.000000 154.630005 155.320007 155.320007 \n", "19 2017-08-11 156.600006 158.570007 156.070007 157.479996 157.479996 \n", "20 2017-08-14 159.320007 160.210007 158.750000 159.850006 159.850006 \n", "\n", " Volume LogAdj \n", "0 23793500 2.173113 \n", "1 17868800 2.174621 \n", "2 20923000 2.177332 \n", "3 17243700 2.175372 \n", "4 26252600 2.175170 \n", "5 21493200 2.180399 \n", "6 18853900 2.182251 \n", "7 15781000 2.184293 \n", "8 32476300 2.176007 \n", "9 17213700 2.172939 \n", "10 19845900 2.170696 \n", "11 35368600 2.174534 \n", "12 69936800 2.194585 \n", "13 27097300 2.190224 \n", "14 20559900 2.192507 \n", "15 21870300 2.199176 \n", "16 36205900 2.202635 \n", "17 26131500 2.205286 \n", "18 40804300 2.191227 \n", "19 26257100 2.197225 \n", "20 22028200 2.203713 " ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import numpy as np\n", "df['LogAdj'] = np.log10(df['Adj Close'])\n", "df" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Para eliminar una columna de nuestro DataFrame usamos el siguiente código:" ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DateOpenHighLowCloseAdj CloseVolume
02017-07-17148.820007150.899994148.570007149.559998148.97497623793500
12017-07-18149.199997150.130005148.669998150.080002149.49295017868800
22017-07-19150.479996151.419998149.949997151.020004150.42927620923000
32017-07-20151.500000151.740005150.190002150.339996149.75192317243700
42017-07-21149.990005150.440002148.880005150.270004149.68220526252600
52017-07-24150.580002152.440002149.899994152.089996151.49507121493200
62017-07-25151.800003153.839996151.800003152.740005152.14254818853900
72017-07-26153.350006153.929993153.059998153.460007152.85972615781000
82017-07-27153.750000153.990005147.300003150.559998149.97106932476300
92017-07-28149.889999150.229996149.190002149.500000148.91520717213700
102017-07-31149.899994150.330002148.130005148.729996148.14822419845900
112017-08-01149.100006150.220001148.410004150.050003149.46305835368600
122017-08-02159.279999159.750000156.160004157.139999156.52533069936800
132017-08-03157.050003157.210007155.020004155.570007154.96147227097300
142017-08-04156.070007157.399994155.690002156.389999155.77825920559900
152017-08-07157.059998158.919998156.669998158.809998158.18879721870300
162017-08-08158.600006161.830002158.270004160.080002159.45382736205900
172017-08-09159.259995161.270004159.110001161.059998160.42999326131500
182017-08-10159.899994160.000000154.630005155.320007155.32000740804300
192017-08-11156.600006158.570007156.070007157.479996157.47999626257100
202017-08-14159.320007160.210007158.750000159.850006159.85000622028200
\n", "
" ], "text/plain": [ " Date Open High Low Close Adj Close \\\n", "0 2017-07-17 148.820007 150.899994 148.570007 149.559998 148.974976 \n", "1 2017-07-18 149.199997 150.130005 148.669998 150.080002 149.492950 \n", "2 2017-07-19 150.479996 151.419998 149.949997 151.020004 150.429276 \n", "3 2017-07-20 151.500000 151.740005 150.190002 150.339996 149.751923 \n", "4 2017-07-21 149.990005 150.440002 148.880005 150.270004 149.682205 \n", "5 2017-07-24 150.580002 152.440002 149.899994 152.089996 151.495071 \n", "6 2017-07-25 151.800003 153.839996 151.800003 152.740005 152.142548 \n", "7 2017-07-26 153.350006 153.929993 153.059998 153.460007 152.859726 \n", "8 2017-07-27 153.750000 153.990005 147.300003 150.559998 149.971069 \n", "9 2017-07-28 149.889999 150.229996 149.190002 149.500000 148.915207 \n", "10 2017-07-31 149.899994 150.330002 148.130005 148.729996 148.148224 \n", "11 2017-08-01 149.100006 150.220001 148.410004 150.050003 149.463058 \n", "12 2017-08-02 159.279999 159.750000 156.160004 157.139999 156.525330 \n", "13 2017-08-03 157.050003 157.210007 155.020004 155.570007 154.961472 \n", "14 2017-08-04 156.070007 157.399994 155.690002 156.389999 155.778259 \n", "15 2017-08-07 157.059998 158.919998 156.669998 158.809998 158.188797 \n", "16 2017-08-08 158.600006 161.830002 158.270004 160.080002 159.453827 \n", "17 2017-08-09 159.259995 161.270004 159.110001 161.059998 160.429993 \n", "18 2017-08-10 159.899994 160.000000 154.630005 155.320007 155.320007 \n", "19 2017-08-11 156.600006 158.570007 156.070007 157.479996 157.479996 \n", "20 2017-08-14 159.320007 160.210007 158.750000 159.850006 159.850006 \n", "\n", " Volume \n", "0 23793500 \n", "1 17868800 \n", "2 20923000 \n", "3 17243700 \n", "4 26252600 \n", "5 21493200 \n", "6 18853900 \n", "7 15781000 \n", "8 32476300 \n", "9 17213700 \n", "10 19845900 \n", "11 35368600 \n", "12 69936800 \n", "13 27097300 \n", "14 20559900 \n", "15 21870300 \n", "16 36205900 \n", "17 26131500 \n", "18 40804300 \n", "19 26257100 \n", "20 22028200 " ] }, "execution_count": 19, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df = df.drop(columns=['LogAdj'])\n", "df" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "

Gráficas\n", "

" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Podemos utilizar gráficas para hacer exploración de datos dentro del DataFrame. Por ejemplo, utilizando la función `plot()` de la librería `matplotlib` creamos un histograma de una variable de interés." ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import matplotlib.pyplot as plt\n", "\n", "# Crear un histograma de la columna 'Adj Close'\n", "df['Adj Close'].plot(kind='hist', bins=5, edgecolor='black')\n", "\n", "# Añadir títulos y etiquetas\n", "plt.title('Histograma del Precio de cierre ajustado')\n", "plt.xlabel('Precio')\n", "plt.ylabel('Frecuencia')\n", "\n", "# Mostrar el gráfico\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "También podemos hacer gráficos de dispersión, en este caso queremos crear un gráfico de dispersión de los precios de cierre versus los precios máximos." ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Crear un gráfico de dispersión\n", "df.plot(kind='scatter', x='Close', y='High', color='blue', edgecolor='black')\n", "\n", "# Añadir títulos y etiquetas\n", "plt.title('Gráfico de Dispersión')\n", "plt.xlabel('Precio de cierre')\n", "plt.ylabel('Precio máximo')\n", "\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "

Ajuste de distribuciones\n", "

" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Para realizar un ajuste de datos a una distribución de probabilidad se puede utilizar la librería de `scipy` con el módulo `stats` y se debe tener una intuición sobre la naturaleza de los datos (continuos o discretos)." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "

Ajuste de distribuciones continuas\n", "

" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Algunas de las distribuciones continuas son: exponencial, Gamma, Weibull, normal, lognormal, uniforme, entre otras. Para conocer si un vector de datos se ajusta a alguna de estas distribuciones puede utilizar la librería de `scipy`. Suponga que el vector `tClientes` contiene el tiempo entre llegadas de clientes a una tienda. Primero, creamos 300 realizaciones de estos tiempos que seguirán una distribución exponencial con tasa 2 clientes por cualquier unidad de tiempo (p. ej. horas):" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [], "source": [ "tClientes = np.random.exponential(scale=1/2, size=300)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Supongamos que no conocemos en un principio qué tipo de distribución siguen los datos del vector `tClientes`. De esta manera, primero haremos un histograma para ver si los datos se comportan de acuerdo a una distribución conocida." ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plt.hist(tClientes, edgecolor='black')\n", "\n", "# Añadir títulos y etiquetas\n", "plt.title('Histograma de la serie de datos')\n", "plt.xlabel('Tiempo entre llegada')\n", "plt.ylabel('Frecuencia')\n", "\n", "# Mostrar el gráfico\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Podemos evidenciar que los datos tienen un comportamiento que se asemeja a la distribución exponencial. Por este motivo, con ayuda de la función ktest() vamos a realizar una prueba de bondad de ajuste por Kolmogorov-Smirnov utilizando la función `kstest()`. Esta recibe como parámetro la distribución que queremos probar. Siendo “expon” la distribución exponencial, \"norm\" la distribución normal, “gamma” la distribución gamma, etc." ] }, { "cell_type": "code", "execution_count": 24, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Estadístico de KS: 0.06757790865099\n", "P-value: 0.12331707428606553\n", "Tasa estimada: 1.7998278910138512\n" ] } ], "source": [ "from scipy import stats\n", "\n", "# Prueba de Kolmogorov-Smirnov\n", "ks_stat, ks_p_value = stats.kstest(tClientes, 'expon',args=(0,1/2))\n", "\n", "print('Estadístico de KS:', ks_stat)\n", "print('P-value:', ks_p_value)\n", "print('Tasa estimada:', 1/(np.mean(tClientes)))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Como pudimos observar, se obtuvo un p-value mayor a 0.05, lo cual nos indica que con una significancia del 5% no rechazamos la hipótesis nula y nuestros datos siguen la distribución exponencial." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Adicionalmente, podemos ver algunas gráficas de interés para la prueba de bondad de ajuste realizada." ] }, { "cell_type": "code", "execution_count": 27, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import statsmodels.api as sm\n", "\n", "# Ajustar la distribución exponencial a los datos\n", "loc, scale = stats.expon.fit(tClientes)\n", "\n", "sm.qqplot(tClientes, dist=stats.expon, loc=loc, scale=scale, line='45')\n", "plt.title('Q-Q plot')\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 28, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "params = stats.expon.fit(tClientes)\n", "\n", "# CDF empírica y teórica\n", "ecdf = sm.distributions.ECDF(tClientes)\n", "x = np.linspace(min(tClientes), max(tClientes))\n", "y = ecdf(x)\n", "plt.plot(x, y, marker='o', linestyle='none')\n", "plt.plot(x, stats.expon.cdf(x, *params), 'r')\n", "plt.title('Empirical and theoretical CDFs')\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 29, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# P-P plot\n", "pp_plot = sm.ProbPlot(tClientes, dist=stats.expon, fit=True)\n", "pp_plot.ppplot(line='45')\n", "plt.title('P-P plot')\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "

Ajuste de distribuciones discretas\n", "

" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Algunas de las distribuciones discretas son: Poisson, geométrica, binomial, hipergeométrica y binomial negativa, entre otras. A continuación generamos un vector de datos que siguen una distribución Poisson con tasa de 3." ] }, { "cell_type": "code", "execution_count": 30, "metadata": {}, "outputs": [], "source": [ "tasa = 3\n", "x2 = np.random.poisson(tasa, 50)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Evaluamos si los datos siguen la distribución Poisson" ] }, { "cell_type": "code", "execution_count": 31, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Valor estadístico: 6.361453362354766\n", "P-value: 0.49823397451165086\n" ] } ], "source": [ "from scipy.stats import poisson, chisquare\n", "\n", "#Calcular la media de la muestra:\n", "media = np.mean(x2)\n", "\n", "#Calcular las frecuencias observadas:\n", "valores, frecuencias_observadas = np.unique(x2, return_counts=True)\n", "\n", "#Calcular las frecuencias esperadas:\n", "frecuencias_esperadas = poisson.pmf(valores, media) * len(x2)\n", "\n", "# Ajuste de las frecuencias esperadas para que sumen lo mismo que las observadas\n", "frecuencias_esperadas = frecuencias_esperadas * frecuencias_observadas.sum() / frecuencias_esperadas.sum()\n", "\n", "#Realizar prueba chi-cuadrado\n", "estadistico_chi2, p_value = chisquare(frecuencias_observadas, frecuencias_esperadas)\n", "\n", "print('Valor estadístico:', estadistico_chi2)\n", "print('P-value:', p_value)\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Como pudimos observar, se obtuvo un p-value mayor a 0.05, lo cual nos indica que con una significancia del 5% no rechazamos la hipótesis nula y nuestros datos siguen la distribución Poisson." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "

Ejercicios\n", "

" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "1. Utilizando el archivo de datos “AAPL.csv”, grafique un histograma del precio de cierre de la acción para los días martes y viernes.\n", "2. Cree una nueva columna que contenga la diferencia entre el precio de apertura y cierre. Luego cree un subconjunto de datos en donde estén las 20 observaciones con mayor valor de la nueva variable." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Universidad de los Andes | Vigilada Mineducación. Reconocimiento como Universidad: Decreto 1297 del 30 de mayo de 1964. Reconocimiento personería jurídica: Resolución 28 del 23 de febrero de 1949 Minjusticia. Departamento de Ingeniería Industrial Carrera 1 Este No. 19 A 40 Bogotá, Colombia Tel. (57.1) 3324320 | (57.1) 3394949 Ext. 2880 /2881 http://industrial.uniandes.edu.co" ] } ], "metadata": { "kernelspec": { "display_name": ".venv", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.13.1" } }, "nbformat": 4, "nbformat_minor": 2 }