a collect of tools for analyzing stock prices


License
MIT
Install
pip install stockpriceanna==1.1.0

Documentation

stockpriceanna

a collection of tools for analyzing stock prices

current version: 1.0.8

installation:

pip install stockpriceanna 

or

pip install stockpriceanna --upgrade

required package: pandas, numpy

Completed projects

VR_test:

This method takes a array-like data as input and perform variance ratio test according to Lo and MacKinlay's algorithem for testing the hypothsis that price follows gemetric random walk

The test detects deviation from log-normal distribution in asset returns by comparing estimated variances from different holding periods

This method allows array-like inputs for the choise of long-period variance and it returns the Z statistics or the p-values corresponding to each input

example:

from stockpriceanna.utils import VR_test
from pandas_datareader import data

p1= data.DataReader('AAPL', 'yahoo', "2013-01-01", "2019-01-01")  #a super large cap stock
VR_test(data=p1["Adj Close"], sub_size=range(2,10))

output:  (do not reject the null)
lvar_size
2    0.3264
3   -0.0111
4   -0.6196
5    0.4911
6   -0.6539
7   -0.3045
8   -0.9522
9    0.2282
Name: Z-stat, dtype: float64

p2= data.DataReader('ACLS', 'yahoo', "2013-01-01", "2019-01-01") #a small cap stock
VR_test(data=p1["Adj Close"], sub_size=range(2,10))
output:  (some evidences for rejecting the null)
lvar_size
2   -1.4314
3   -1.4439
4   -2.3415
5   -2.0108
6   -2.9587
7   -2.6798
8   -1.9327
9   -2.7329
Name: Z-stat, dtype: float64

pricegen

The pricegen class generates a simulated price series made of two components - the real and the noise. The noise component has only an one-period effect on price while the real component has a permanent effect on price. User may add customized behaviors to the artifical stock prices including: determinsitic trend, decaying effect of random shock on stock prices, autocorrelation, and price momentum. The generated factor and noise components are stored in the self.data dataframe for analysis from an insider's view.

example:

from stockpriceanna.pricegen import spgen
import numpy as np

d1 = spgen(size=1000)  #specify the size of the simulation
d1.gen_ln(mean=0,var=0.01,seed=1,name="factor") #create a log-normal component as a factor
d1.gen_ln(mean=0,var=0.005,name="noise",add_type="noise",seed=2) #create a log-normal component as a noise
d1.gen_ac(base="factor",rate=0.2,lag=3,name="lag3_ac") #specify how previous shock to stock price decay overtime
d1.gen_trend(rate=0.0005,name="trend") #set a trend 
d1.gen_custom(func =lambda:(1+ 0.02*(np.random.rand()-0.5)),name="unit_noise",add_type="noise") #create another noise variable that follows unitform distribution
d1.gen_custom(func = lambda l1,l2: (l1+l2)/2 if (l1-1)*(l2-1)>0 else 1,var="unit_noise",name="mom_factor",add_type="factor") #create a momentum factor based on the lag values of the uniform noise
d1.gen_price(method="m") #generate the stock prices using multiplicative model
d1.gen_periodic(name="2period_factor",func=lambda p,var: 1.1*var if p==0 else 0.9*var,add_type="noise",period=2,var="noise") #generates a periodic filter that amplifier/reduce the base noise by 10% in cycle. 
d1.gen_teffect(name="Jan_effect",func=lambda t: 1.001 if t.month==1 else 1,add_type="factor") #generates a January effect that boosts stock returns by a daily factor.

d1.show_price(show_rprice=True).tail(10)

output:
                 price     price_r
2019-05-07  206.818631  207.715480
2019-05-08  208.517919  207.593980
2019-05-09  209.955386  208.456761
2019-05-10  211.557670  211.442844
2019-05-11  213.869664  215.467602
2019-05-12  216.946547  215.952856
2019-05-13  212.117639  211.268894
2019-05-14  212.944435  211.194571
2019-05-15  211.314871  213.129797
2019-05-16  213.727595  212.944717
"""