In [1]:

    

from IPython.display import HTML
HTML('''<script> code_show=true;  
function code_toggle() {
 if (code_show){  $('div.input').hide(); 
 } else {  $('div.input').show();  }
 code_show = !code_show
}
$( document ).ready(code_toggle); </script>
The raw code for this IPython notebook is by default hidden for easier reading.
To toggle on/off the raw code, click <a href="javascript:code_toggle()">here</a>.''')


    

    
Out[1]:





The raw code for this IPython notebook is by default hidden for easier reading.
To toggle on/off the raw code, click here.

In [2]:

    

import pandas as pd
import numpy as np
import seaborn as sns


    

In [3]:

    

%matplotlib inline


    

In [4]:

    

def read_stock(stock_name):
    fname = ("table_%s.csv" % stock_name)
#     print(fname)
    data1 = pd.read_csv(fname, names=['date', 'foo', 'o', 'h', 'l', 'c', 'v'], header=False, parse_dates=['date'])
    return data1


    

In [5]:

    

read_stock("aapl")['o'].plot()


    

    
Out[5]:





<matplotlib.axes._subplots.AxesSubplot at 0x1122fbba8>






    

In [174]:

    

def compute_scenarios(stock, λ = 0.06):
    _com = (1 - λ) / λ
    
    log_returns = np.log( stock / stock.shift(1))
    log_returns = log_returns.ix[1:]
    
    _std = pd.ewmstd( log_returns, com=_com)
    _mu = pd.ewma( log_returns, com=_com)

    residuals = ( log_returns / _std )
    residuals = residuals.ix[20:]

    σ = _std[_std.last_valid_index()]
    scenario_shifts = np.exp(residuals * σ)
    neutral_scenario = stock.ix[stock.last_valid_index()]
    scenarios = neutral_scenario * scenario_shifts
    
    return scenarios, neutral_scenario, scenario_shifts


def compute_logreturns_np(stock):
    return np.log( np.divide(stock[1:], stock[:-1]) )

def compute_scenarios_np(stock, λ = 0.06, number_of_scenarios=752):
    _com = (1 - λ) / λ
    
    last_index = len(stock) - 1
    first_index = last_index - number_of_scenarios
    log_returns = compute_logreturns_np(stock)
    
    _std = pd.ewmstd( log_returns, com=_com, ignore_na=True)
    _std = _std[first_index:last_index]
    _mu  = pd.ewma( stock, com=_com, ignore_na=True)
    _mu  = _mu[first_index:last_index]

    ξ = np.divide(log_returns[first_index:last_index], _std)
    neutral_scenario, σ = stock[-1], _std[-1]
    scenario_shifts = np.exp(ξ * σ)
    scenarios = neutral_scenario * scenario_shifts
    
    return scenarios, neutral_scenario, log_returns[-number_of_scenarios:], _std, _mu


def compute_var(scenarios, neutral_scenario, level=99.):
    pnls = scenarios - neutral_scenario
    return - np.percentile(pnls, 100-level, interpolation='linear'), pnls

def f(stock, λ=0.06):
    scenarios, neutral, log_returns, σ = compute_scenarios_np(stock)
    var = compute_var(scenarios, neutral)
    sns.distplot(neutral - scenarios);
    return np.mean(scenarios)

def plot_all(scenarios, neutral_scenario, log_returns, σ, ξ=None):
#     com = (1 - _lambda) / _lambda
    ig, axes = sns.plt.subplots(nrows=4, ncols=1, figsize=(12,8))
    axes[0].set_title('Log returns')
    sns.distplot( log_returns,ax=axes[0]) #, label='log-returns')
    sns.distplot( np.exp(log_returns), ax=axes[1] )
    axes[2].set_title('Estimated volatility')
    sns.tsplot( σ, legend=True, ax=axes[2]) #, label='std')
#     sns.tsplot( scenarios-neutral, legend=True, ax=axes[3]) #, label='std')
    sns.distplot( scenarios, ax=axes[3] )

def f1(stock, λ=0.06):
    scenarios, neutral, log_returns, σ, μ = compute_scenarios_np(stock)
#     plot_all(scenarios, neutral, log_returns, σ)
    var = compute_var(scenarios, neutral)
#     sns.distplot(neutral - scenarios);
    return var


    

In [9]:

    

stock = read_stock("aapl")['o']
scenarios, neutral, shifts = compute_scenarios(stock)
sns.distplot(neutral-scenarios);


    

In [14]:

    

from IPython.html.widgets import *
def g(x):
    var = f1(stock.ix[x-1000:x].values)
#     print(stock.ix[x], var)
interact(g, x=(1001,3924,10));


    

In [15]:

    

from bokeh.plotting import *


    

In [16]:

    

output_notebook()


    

    





    
        
        
        
    

        
        BokehJS successfully loaded.
    
    
Warning: BokehJS previously loaded

In [184]:

    

x = np.arange(752)
y = np.random.rand(752)
hist, edges = np.histogram(scenarios, density=True, bins=50)

source = ColumnDataSource(data=dict(x=x, y=y,
                                    hist=hist, edges=edges,
                                    xx=np.append(x,x[::-1]), yy=np.append(y, y[::-1]),
                                   x_var = 0, var = 0))


@interact(x=(1001,3924,10), λ=(0.01, 0.99, 0.01), α=(0.01, 0.99, 0.01))
def update(x, λ, α):
    s = stock.ix[x-1000:x].values
    scenarios, neutral, log_returns, σ, μ = compute_scenarios_np(s, λ=λ)
    source.data['y'] = scenarios - neutral
    var = compute_var(scenarios, neutral, level=α)
    
    hist, edges = np.histogram(scenarios - neutral, density=True, bins=50)
    
    source.data['top'] = hist
    source.data['left'] = edges[:-1]
    source.data['right'] = edges[1:]
    source.data['bottom'] = 0
    source.data['xvar'] = [source.data['x'][0], source.data['x'][-1]]
    source.data['var'] = var
#     print(var)
    
    source.data['yy'] = np.append( np.multiply(s[-752:], np.exp(1 - σ)), np.multiply(s[-752:], 1 + σ)[::-1])
    source.push_notebook()

p = figure(title="simple line example", plot_height=300, plot_width=600)
# p.line('x', 'y', color="#2222aa", line_width=1, source=source, name="foo")
p.patch('xx', 'yy', source=source, color='#7570B3', fill_alpha=0.2)
p.line('x_var', 'var', source=source, line_width=1, color='#444444', name="var")

def stylize(p):
    p.title_text_font_size = "8pt"
    p.axis.major_tick_line_color = None
    p.axis[0].ticker.num_minor_ticks = 0
    p.axis[1].ticker.num_minor_ticks = 0


p1 = figure(title="Returns Histogram", plot_height=300, plot_width=600)
p1.quad(top=[0], bottom=0, left=[0], right=[0],
        fill_color="#036564", line_color="#033649",
        alpha=0.5,
        source=source)



stylize(p)
stylize(p1)


    

In [185]:

    

show(gridplot([[p],[p1]]))


    

In [21]:

    

%%timeit
scenarios, neutral, shifts = compute_scenarios(stock)
var = compute_var(scenarios, neutral)


    

    




100 loops, best of 3: 2.57 ms per loop

In [180]:

    

df = pd.DataFrame()
idx = 1
def accumul4te(stock, λ=0.06):
    global idx
    scenarios, neutral, log_returns, σ, μ = compute_scenarios_np(stock, λ=λ)
#     print(scenarios)
    df[str(idx)] = scenarios
    idx = idx + 1
    var, pnls = compute_var(scenarios, neutral)
    return var #np.mean(pnls)


    

In [186]:

    

# %%timeit
hist_var = pd.stats.moments.rolling_apply( stock, 1000, accumul4te, min_periods=850)
#


    

In [189]:

    

sns.tsplot(hist_var)


    

    
Out[189]:





<matplotlib.axes._subplots.AxesSubplot at 0x11d1a4160>






    

In [195]:

    

# fig = figure(title="simple line example", plot_height=600, plot_width=600)
x = np.zeros(0)
y = np.zeros(0)
for c in np.arange(0,3600, 10):
    col = df.columns[c]
    x = np.append(int(col) * np.ones(752), x)
    y = np.append(df[col].values - hist_var[int(col)], y)


    

In [196]:

    

# ax = sns.plt.gca()
fig, ax = sns.plt.subplots(1)
fig.set_size_inches((17,8))
ax.scatter(x, y)


    

    
Out[196]:





<matplotlib.collections.PathCollection at 0x122b6b5c0>






    

In [ ]:

    

y - hist_var


    

In [98]:

    

x


    

    
Out[98]:





array([], dtype=float64)

In [71]:

    

x


    

    
Out[71]:





array([ 10.,  10.,  10., ...,   1.,   1.,   1.])

In [150]:

    

upperband = np.random.random_integers(100, 150, size=100)
lowerband = upperband - 100
x_data = np.arange(1, 101)

band_x = np.append(x_data, x_data[::-1]) 
band_y = np.append(lowerband, upperband[::-1])


    

In [158]:

    

sns.plt.plot(band_x, band_y)


    

    
Out[158]:





[<matplotlib.lines.Line2D at 0x1199e0cf8>]






    

In [281]:

    

sns.plt.plot(stock)


    

    
Out[281]:





[<matplotlib.lines.Line2D at 0x11ee48be0>]






    

In [135]:

    

p.patch?


    

In [129]:

    

from IPython.core.display import HTML
import requests
def css_styling():
    styles = requests.get('http://nbviewer.ipython.org/github/barbagroup/CFDPython/blob/master/styles/custom.css').text
    print(styles)
    return HTML(styles)
css_styling()


    

    




<link href='http://fonts.googleapis.com/css?family=Fenix' rel='stylesheet' type='text/css'>
<link href='http://fonts.googleapis.com/css?family=Alegreya+Sans:100,300,400,500,700,800,900,100italic,300italic,400italic,500italic,700italic,800italic,900italic' rel='stylesheet' type='text/css'>
<link href='http://fonts.googleapis.com/css?family=Source+Code+Pro:300,400' rel='stylesheet' type='text/css'>
<style>
    @font-face {
        font-family: "Computer Modern";
        src: url('http://mirrors.ctan.org/fonts/cm-unicode/fonts/otf/cmunss.otf');
    }
    div.cell{
        width:800px;
        margin-left:16% !important;
        margin-right:auto;
    }
    h1 {
        font-family: 'Alegreya Sans', sans-serif;
    }
    h2 {
        font-family: 'Fenix', serif;
    }
    h3{
		font-family: 'Fenix', serif;
        margin-top:12px;
        margin-bottom: 3px;
       }
	h4{
		font-family: 'Fenix', serif;
       }
    h5 {
        font-family: 'Alegreya Sans', sans-serif;
    }	   
    div.text_cell_render{
        font-family: 'Alegreya Sans',Computer Modern, "Helvetica Neue", Arial, Helvetica, Geneva, sans-serif;
        line-height: 135%;
        font-size: 120%;
        width:600px;
        margin-left:auto;
        margin-right:auto;
    }
    .CodeMirror{
            font-family: "Source Code Pro";
			font-size: 90%;
    }
/*    .prompt{
        display: None;
    }*/
    .text_cell_render h1 {
        font-weight: 200;
        font-size: 50pt;
		line-height: 100%;
        color:#CD2305;
        margin-bottom: 0.5em;
        margin-top: 0.5em;
        display: block;
    }	
    .text_cell_render h5 {
        font-weight: 300;
        font-size: 16pt;
        color: #CD2305;
        font-style: italic;
        margin-bottom: .5em;
        margin-top: 0.5em;
        display: block;
    }
    
    .warning{
        color: rgb( 240, 20, 20 )
        }  
</style>
<script>
    MathJax.Hub.Config({
                        TeX: {
                           extensions: ["AMSmath.js"]
                           },
                tex2jax: {
                    inlineMath: [ ['$','$'], ["\\(","\\)"] ],
                    displayMath: [ ['$$','$$'], ["\\[","\\]"] ]
                },
                displayAlign: 'center', // Change this to 'center' to center equations.
                "HTML-CSS": {
                    styles: {'.MathJax_Display': {"margin": 4}}
                }
        });
</script>







    
Out[129]: