├── .idea
    └── vcs.xml
├── LICENSE
├── README.md
├── in
    └── IBM_cts.h5
├── main.py
└── out
    └── realized_quantities_IBM_cts.csv


/.idea/vcs.xml:
--------------------------------------------------------------------------------
1 | <?xml version="1.0" encoding="UTF-8"?>
2 | <project version="4">
3 |   <component name="VcsDirectoryMappings">
4 |     <mapping directory="$PROJECT_DIR$" vcs="Git" />
5 |   </component>
6 | </project>


--------------------------------------------------------------------------------
/LICENSE:
--------------------------------------------------------------------------------
 1 | MIT License
 2 | 
 3 | Copyright (c) 2019 Sebastian Bayer
 4 | 
 5 | Permission is hereby granted, free of charge, to any person obtaining a copy
 6 | of this software and associated documentation files (the "Software"), to deal
 7 | in the Software without restriction, including without limitation the rights
 8 | to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
 9 | copies of the Software, and to permit persons to whom the Software is
10 | furnished to do so, subject to the following conditions:
11 | 
12 | The above copyright notice and this permission notice shall be included in all
13 | copies or substantial portions of the Software.
14 | 
15 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17 | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
18 | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19 | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
20 | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
21 | SOFTWARE.
22 | 


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
 1 | # Realized Quantities
 2 | 
 3 | The main.py file contains a large variety of realized quantities.
 4 | To date, the following are implemented:
 5 | 
 6 | * Realized Variance
 7 | * Realized Semivariance
 8 | * Realized Skewness
 9 | * Realized Kurtosis
10 | * Realized Absolute Variation
11 | * Realized Bipower Variation
12 | * Standardized Tri-Power Quarticity
13 | * Signed Jump Variation
14 | 
15 | This file uses the data generated from the
16 | [ResampleHF](https://github.com/BayerSe/ResampleHF) repo.


--------------------------------------------------------------------------------
/in/IBM_cts.h5:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/BayerSe/RealizedQuantities/663b81ffb82d11315012a08076b63a2a0485a879/in/IBM_cts.h5


--------------------------------------------------------------------------------
/main.py:
--------------------------------------------------------------------------------
 1 | #!/usr/bin/env python
 2 | 
 3 | import numpy as np
 4 | import pandas as pd
 5 | from scipy import stats
 6 | from scipy.special import gamma
 7 | 
 8 | 
 9 | def realized_quantity(fun):
10 |     """Applies the function 'fun' to each day separately"""
11 |     return intraday_returns.groupby(pd.Grouper(freq="1d")).apply(fun)[index]
12 | 
13 | 
14 | if __name__ == "__main__":
15 | 
16 |     # Settings
17 |     asset = "IBM"
18 |     sampling = "cts"
19 |     trading_seconds = 23400
20 |     avg_sampling_frequency = 300
21 |     original_sampling_frequency = 60  # From the process_data.py file
22 |     M = trading_seconds / original_sampling_frequency
23 | 
24 |     # Load data and store the intraday returns
25 |     data = pd.read_hdf("in/" + asset + "_" + sampling + ".h5", "table")
26 |     intraday_returns = data.groupby(pd.Grouper(freq="1d")).apply(lambda x: np.log(x / x.shift(1))).dropna()
27 | 
28 |     # Index of all days
29 |     index = data.groupby(pd.Grouper(freq="1d")).first().dropna().index
30 | 
31 |     # Some constants
32 |     mu_1 = np.sqrt((2 / np.pi))
33 |     mu_43 = 2 ** (2 / 3) * gamma(7 / 6) * gamma(1 / 2) ** (-1)
34 | 
35 |     # First and last price of each day
36 |     prices_open = data.resample('D').first()[index]
37 |     prices_close = data.resample('D').last()[index]
38 | 
39 |     # Return (close-to-close and open-to-close
40 |     r_cc = pd.Series(np.log(prices_close / prices_close.shift(1)), name='r_cc')
41 |     r_oc = pd.Series(np.log(prices_close / prices_open), name='r_oc')
42 | 
43 |     # Realized Variance (Andersen and Bollerslev, 1998)
44 |     rv = realized_quantity(lambda x: (x ** 2).sum())
45 | 
46 |     # Realized absolute variation (Forsberg and Ghysels, 2007)
47 |     rav = mu_1 ** (-1) * M ** (-.5) * realized_quantity(lambda x: x.abs().sum())
48 | 
49 |     # Realized bipower variation (Barndorff-Nielsen and Shephard; 2004, 2006)
50 |     bv = mu_1 ** (-2) * realized_quantity(lambda x: (x.abs() * x.shift(1).abs()).sum())
51 | 
52 |     # Standardized tri-power quarticity (see e.g. Forsberg & Ghysels, 2007)
53 |     tq = M * mu_43 ** (-3) * realized_quantity(
54 |         lambda x: (x.abs() ** (4 / 3) * x.shift(1).abs() ** (4 / 3) * x.shift(2).abs() ** (4 / 3)).sum())
55 | 
56 |     # Jump test by Huang and Tauchen (2005)
57 |     j = (np.log(rv) - np.log(bv)) / \
58 |         ((mu_1 ** -4 + 2 * mu_1 ** -2 - 5) / (M * tq * bv ** -2)) ** 0.5
59 |     jump = j.abs() >= stats.norm.ppf(0.999)
60 | 
61 |     # Separate continuous and discontinuous parts of the quadratic variation
62 |     iv = pd.Series(0, index=index)
63 |     iv[jump] = bv[jump] ** 0.5
64 |     iv[~jump] = rv[~jump] ** 0.5
65 | 
66 |     jv = pd.Series(0, index=index)
67 |     jv[jump] = rv[jump] ** 0.5 - bv[jump] ** 0.5
68 |     jv[jv < 0] = 0
69 | 
70 |     # Realized Semivariance (Barndorff-Nielsen, Kinnebrock and Shephard, 2010)
71 |     rv_m = realized_quantity(lambda x: (x ** 2 * (x < 0)).sum())
72 |     rv_p = realized_quantity(lambda x: (x ** 2 * (x > 0)).sum())
73 | 
74 |     # Signed jump variation (Patton and Sheppard, 2015)
75 |     sjv = rv_p - rv_m
76 |     sjv_p = sjv * (sjv > 0)
77 |     sjv_m = sjv * (sjv < 0)
78 | 
79 |     # Realized Skewness and Kurtosis  (see, e.g. Amaya, Christoffersen, Jacobs and Vasquez, 2015)
80 |     rm3 = realized_quantity(lambda x: (x ** 3).sum())
81 |     rm4 = realized_quantity(lambda x: (x ** 4).sum())
82 |     rs = np.sqrt(M) * rm3 / rv ** (3 / 2)
83 |     rk = M * rm4 / rv ** 2
84 | 
85 |     # Export data
86 |     out = pd.concat([r_cc, r_oc, rav, rv ** .5, bv ** .5, rv_m ** 0.5, rv_p ** 0.5,
87 |                      iv, jv, sjv, sjv_p, sjv_m, rs, rk], axis=1)
88 |     out.columns = ['r_cc', 'r_oc', 'rav', 'rvol', 'bvol', 'rvol_m', 'rvol_p',
89 |                    'ivol', 'jvol', 'sjv', 'sjv_p', 'sjv_m', 'rs', 'rk']
90 |     out.to_csv('out/realized_quantities_' + asset + "_" + sampling + ".csv")
91 | 


--------------------------------------------------------------------------------