Leer los datos¶

In [26]:

import pandas as pd
import matplotlib.pyplot as plt
url='https://raw.githubusercontent.com/JJTorresDS/ds-data-sources/main/stocks.csv'
df= pd.read_csv(url,index_col='formatted_date',parse_dates=['formatted_date'])
df

Out[26]:

	MCD	SBUX	GOOG	AMZN	MSFT	JPM	BAC	C	MAR	HLT	RCL	V	MA	PYPL
formatted_date
2016-01-01	106.332146	54.353962	742.950012	587.000000	49.853489	50.424938	12.573010	36.897804	57.754189	35.192841	74.235298	71.574371	85.822624	36.139999
2016-02-01	100.671043	52.064243	697.770020	552.520020	46.043667	48.033066	11.132540	33.707108	64.228912	41.061607	67.360649	69.556580	83.956566	38.139999
2016-03-01	108.782211	53.571442	744.950012	593.640015	50.339031	50.524323	12.021718	36.223217	67.336624	44.499886	74.790009	73.631477	91.278160	38.599998
2016-04-01	109.483307	50.457645	693.010010	659.590027	45.453705	53.919910	12.995729	40.153545	66.305466	43.716049	70.465584	74.363144	93.683258	39.180000
2016-05-01	105.648926	49.255203	735.719971	722.789978	48.306515	56.098225	13.201019	40.447887	62.474155	41.198154	70.456474	75.999847	92.817329	37.790001
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
2021-07-01	240.055786	120.441193	2704.419922	3327.590088	283.846313	150.006454	37.981613	66.635887	145.979996	131.449997	76.870003	245.622177	384.999420	275.529999
2021-08-01	234.863205	116.533272	2909.239990	3470.790039	300.752960	158.991928	41.338173	71.392593	135.139999	124.860001	82.730003	228.386063	345.793121	288.660004
2021-09-01	239.780670	109.836403	2665.310059	3285.040039	281.405182	162.709534	42.031269	69.675041	148.089996	132.110001	88.949997	222.358551	347.241272	260.209991
2021-10-01	244.196198	105.614609	2965.409912	3372.429932	331.014404	168.872391	47.551128	68.662384	160.020004	143.949997	84.430000	211.397858	335.096588	232.589996
2021-11-01	243.251434	109.169281	2849.040039	3507.070068	329.986298	158.830002	44.256985	63.700001	147.559998	135.070007	69.820000	193.429489	314.920013	184.889999

71 rows × 14 columns

In [27]:

df.corr()

Out[27]:

	MCD	SBUX	GOOG	AMZN	MSFT	JPM	BAC	C	MAR	HLT	RCL	V	MA	PYPL
MCD	1.000000	0.911649	0.872950	0.915207	0.921861	0.900587	0.857553	0.598726	0.758567	0.941815	0.052275	0.961073	0.956230	0.889973
SBUX	0.911649	1.000000	0.900343	0.874861	0.925871	0.854682	0.775915	0.463075	0.608303	0.895421	-0.138369	0.908692	0.907381	0.921438
GOOG	0.872950	0.900343	1.000000	0.898136	0.961937	0.909442	0.876196	0.530359	0.647768	0.905074	-0.130685	0.855338	0.841940	0.927271
AMZN	0.915207	0.874861	0.898136	1.000000	0.963040	0.831325	0.773567	0.414525	0.596446	0.859951	-0.207798	0.948422	0.951015	0.959953
MSFT	0.921861	0.925871	0.961937	0.963040	1.000000	0.873916	0.818282	0.436713	0.596887	0.897585	-0.232318	0.929751	0.926621	0.951394
JPM	0.900587	0.854682	0.909442	0.831325	0.873916	1.000000	0.982349	0.794576	0.872046	0.978772	0.202581	0.871820	0.859420	0.844586
BAC	0.857553	0.775915	0.876196	0.773567	0.818282	0.982349	1.000000	0.841762	0.899805	0.950060	0.292480	0.805341	0.789793	0.773794
C	0.598726	0.463075	0.530359	0.414525	0.436713	0.794576	0.841762	1.000000	0.913796	0.740547	0.686983	0.531859	0.510164	0.439464
MAR	0.758567	0.608303	0.647768	0.596446	0.596887	0.872046	0.899805	0.913796	1.000000	0.871625	0.581135	0.692143	0.679711	0.580950
HLT	0.941815	0.895421	0.905074	0.859951	0.897585	0.978772	0.950060	0.740547	0.871625	1.000000	0.170140	0.912165	0.905176	0.862439
RCL	0.052275	-0.138369	-0.130685	-0.207798	-0.232318	0.202581	0.292480	0.686983	0.581135	0.170140	1.000000	-0.056921	-0.070195	-0.203879
V	0.961073	0.908692	0.855338	0.948422	0.929751	0.871820	0.805341	0.531859	0.692143	0.912165	-0.056921	1.000000	0.996748	0.921974
MA	0.956230	0.907381	0.841940	0.951015	0.926621	0.859420	0.789793	0.510164	0.679711	0.905176	-0.070195	0.996748	1.000000	0.918581
PYPL	0.889973	0.921438	0.927271	0.959953	0.951394	0.844586	0.773794	0.439464	0.580950	0.862439	-0.203879	0.921974	0.918581	1.000000

Parte 1: Correlaciones¶

In [28]:

import seaborn as sns 
sns.heatmap(df.corr(), cmap='jet')

Out[28]:

<matplotlib.axes._subplots.AxesSubplot at 0x7fbcf009a810>

Parte 2: Analizar si hay valores faltantes¶

In [14]:

df.isnull().sum()

Out[14]:

MCD     0
SBUX    0
GOOG    0
AMZN    0
MSFT    0
JPM     0
BAC     0
C       0
MAR     0
HLT     0
RCL     0
V       0
MA      0
PYPL    0
dtype: int64

Para este caso no tenemos problema con los valores faltantes

Parte 3¶

In [19]:

# graficando la serie de tiempo
df.plot(kind='line', figsize=(12,8), xlabel='Fecha', ylabel='Precio', title='Acciones en el tiempo')

Out[19]:

<matplotlib.axes._subplots.AxesSubplot at 0x7fbcf0f5bd10>

In [29]:

df_melt=pd.melt(df)
df_melt

Out[29]:

	variable	value
0	MCD	106.332146
1	MCD	100.671043
2	MCD	108.782211
3	MCD	109.483307
4	MCD	105.648926
...	...	...
989	PYPL	275.529999
990	PYPL	288.660004
991	PYPL	260.209991
992	PYPL	232.589996
993	PYPL	184.889999

994 rows × 2 columns

In [23]:

# Con atipicos
plt.figure(figsize=(12,6))
sns.boxplot(data=df)

Out[23]:

<matplotlib.axes._subplots.AxesSubplot at 0x7fbcf03f1790>

In [24]:

# Sin atipicos
plt.figure(figsize=(12,6))
sns.boxplot(data=df, showfliers=False)

Out[24]:

<matplotlib.axes._subplots.AxesSubplot at 0x7fbcf0e90cd0>

In [36]:

# Cambio porcentual
df.pct_change().plot(kind='line', figsize=(15,6),xlabel='Fecha',ylabel='Cambio %')
plt.legend(fontsize='xx-large', ncol=4, loc='lower left')

Out[36]:

<matplotlib.legend.Legend at 0x7fbcef905f90>

Created in Deepnote