![\"Python](\"python_for_finance.png\")
\n",
"\n",
"\n",
"\n",
" \n",
"
\n",
"
"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 4,
"text": [
" Open High Low Close Volume Adj Close\n",
"Date \n",
"2014-03-05 9562.39 9599.00 9534.43 9542.02 73341700 9542.02\n",
"2014-03-06 9577.35 9587.44 9505.32 9542.87 103682600 9542.87\n",
"2014-03-07 9538.45 9543.24 9346.82 9350.75 103246700 9350.75\n",
"2014-03-10 9305.51 9382.98 9216.07 9265.50 84875400 9265.50\n",
"2014-03-11 9295.32 9375.29 9259.18 9307.79 72300800 9307.79\n",
"\n",
"[5 rows x 6 columns]"
]
}
],
"prompt_number": 4
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"pandas makes it easy to implement **vectorized operations**, like calculating log-returns over whole time series."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"index['Returns'] = np.log(index['Close'] / index['Close'].shift(1))"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 5
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"In addition, pandas makes **plotting** quite simple and compact."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"index[['Close', 'Returns']].plot(subplots=True, style='b', figsize=(8, 5))"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 6,
"text": [
"array([| \n", " | Open | \n", "High | \n", "Low | \n", "Close | \n", "Volume | \n", "Adj Close | \n", "
|---|---|---|---|---|---|---|
| Date | \n", "\n", " | \n", " | \n", " | \n", " | \n", " | \n", " |
| 2014-03-05 | \n", "9562.39 | \n", "9599.00 | \n", "9534.43 | \n", "9542.02 | \n", "73341700 | \n", "9542.02 | \n", "
| 2014-03-06 | \n", "9577.35 | \n", "9587.44 | \n", "9505.32 | \n", "9542.87 | \n", "103682600 | \n", "9542.87 | \n", "
| 2014-03-07 | \n", "9538.45 | \n", "9543.24 | \n", "9346.82 | \n", "9350.75 | \n", "103246700 | \n", "9350.75 | \n", "
| 2014-03-10 | \n", "9305.51 | \n", "9382.98 | \n", "9216.07 | \n", "9265.50 | \n", "84875400 | \n", "9265.50 | \n", "
| 2014-03-11 | \n", "9295.32 | \n", "9375.29 | \n", "9259.18 | \n", "9307.79 | \n", "72300800 | \n", "9307.79 | \n", "
5 rows \u00d7 6 columns
\n", "\n",
"\n",
" \n",
"
\n",
"
"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 21,
"text": [
" EUROSTOXX VSTOXX\n",
"date \n",
"2000-01-03 4849.22 30.9845\n",
"2000-01-04 4657.83 33.2225\n",
"2000-01-05 4541.75 32.5944\n",
"2000-01-06 4500.69 31.1811\n",
"2000-01-07 4648.27 27.4407\n",
"\n",
"[5 rows x 2 columns]"
]
}
],
"prompt_number": 21
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"A **picture** can tell almost the complete story."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"data.plot(subplots=True, grid=True, style='b', figsize=(10, 5))"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 22,
"text": [
"array([| \n", " | EUROSTOXX | \n", "VSTOXX | \n", "
|---|---|---|
| date | \n", "\n", " | \n", " |
| 2000-01-03 | \n", "4849.22 | \n", "30.9845 | \n", "
| 2000-01-04 | \n", "4657.83 | \n", "33.2225 | \n", "
| 2000-01-05 | \n", "4541.75 | \n", "32.5944 | \n", "
| 2000-01-06 | \n", "4500.69 | \n", "31.1811 | \n", "
| 2000-01-07 | \n", "4648.27 | \n", "27.4407 | \n", "
5 rows \u00d7 2 columns
\n", "\n",
"\n",
" \n",
"
\n",
"
"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 23,
"text": [
" EUROSTOXX VSTOXX\n",
"date \n",
"2000-01-03 NaN NaN\n",
"2000-01-04 -0.040268 0.069740\n",
"2000-01-05 -0.025237 -0.019087\n",
"2000-01-06 -0.009082 -0.044328\n",
"2000-01-07 0.032264 -0.127785\n",
"\n",
"[5 rows x 2 columns]"
]
}
],
"prompt_number": 23
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"To this new data set, also stored in a DataFrame object, we apply **OLS**."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"xdat = rets['EUROSTOXX']\n",
"ydat = rets['VSTOXX']\n",
"model = pd.ols(y=ydat, x=xdat)\n",
"model"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 24,
"text": [
"\n",
"-------------------------Summary of Regression Analysis-------------------------\n",
"\n",
"Formula: Y ~ | \n", " | EUROSTOXX | \n", "VSTOXX | \n", "
|---|---|---|
| date | \n", "\n", " | \n", " |
| 2000-01-03 | \n", "NaN | \n", "NaN | \n", "
| 2000-01-04 | \n", "-0.040268 | \n", "0.069740 | \n", "
| 2000-01-05 | \n", "-0.025237 | \n", "-0.019087 | \n", "
| 2000-01-06 | \n", "-0.009082 | \n", "-0.044328 | \n", "
| 2000-01-07 | \n", "0.032264 | \n", "-0.127785 | \n", "
5 rows \u00d7 2 columns
\n", "\n",
"\n",
" \n",
"
\n",
"
"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 29,
"text": [
" EUROSTOXX VSTOXX\n",
"date \n",
"2000-01-03 100.000000 100.000000\n",
"2000-01-04 96.053180 107.222966\n",
"2000-01-05 93.659393 105.195824\n",
"2000-01-06 92.812659 100.634511\n",
"2000-01-07 95.856035 88.562668\n",
"\n",
"[5 rows x 2 columns]"
]
}
],
"prompt_number": 29
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"First, the **initial invest**."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"invest = 100\n",
"cratio = 0.3\n",
"data['Equity'] = (1 - cratio) * invest / data['EUROSTOXX'][0]\n",
"data['Volatility'] = cratio * invest / data['VSTOXX'][0]"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 30
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"This can already be considered an **static** investment strategy."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"data['Static'] = (data['Equity'] * data['EUROSTOXX']\n",
" + data['Volatility'] * data['VSTOXX'])"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 31
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"data[['EUROSTOXX', 'Static']].plot(figsize=(10, 5))"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 32,
"text": [
"| \n", " | EUROSTOXX | \n", "VSTOXX | \n", "
|---|---|---|
| date | \n", "\n", " | \n", " |
| 2000-01-03 | \n", "100.000000 | \n", "100.000000 | \n", "
| 2000-01-04 | \n", "96.053180 | \n", "107.222966 | \n", "
| 2000-01-05 | \n", "93.659393 | \n", "105.195824 | \n", "
| 2000-01-06 | \n", "92.812659 | \n", "100.634511 | \n", "
| 2000-01-07 | \n", "95.856035 | \n", "88.562668 | \n", "
5 rows \u00d7 2 columns
\n", "\n",
"\n",
" \n",
"
\n",
"
"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 35,
"text": [
" EUROSTOXX VSTOXX Equity Volatility Static \\\n",
"date \n",
"2000-01-03 100.000000 100.000000 0.700000 0.300000 100.000000 \n",
"2000-01-04 96.053180 107.222966 0.724420 0.278124 99.404116 \n",
"2000-01-05 93.659393 105.195824 0.725761 0.276930 97.120322 \n",
"2000-01-06 92.812659 100.634511 0.718221 0.283884 95.159214 \n",
"2000-01-07 95.856035 88.562668 0.686354 0.318376 93.668025 \n",
"\n",
" Dynamic \n",
"date \n",
"2000-01-03 100.000000 \n",
"2000-01-04 99.404116 \n",
"2000-01-05 97.106211 \n",
"2000-01-06 95.228521 \n",
"2000-01-07 93.987330 \n",
"\n",
"[5 rows x 6 columns]"
]
}
],
"prompt_number": 35
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"A brief check if the **ratios are indeed constant**."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"(data['Volatility'] * data['VSTOXX'] / data['Dynamic'])[:5]"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 36,
"text": [
"date\n",
"2000-01-03 0.3\n",
"2000-01-04 0.3\n",
"2000-01-05 0.3\n",
"2000-01-06 0.3\n",
"2000-01-07 0.3\n",
"dtype: float64"
]
}
],
"prompt_number": 36
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"(data['Equity'] * data['EUROSTOXX'] / data['Dynamic'])[:5]"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 37,
"text": [
"date\n",
"2000-01-03 0.7\n",
"2000-01-04 0.7\n",
"2000-01-05 0.7\n",
"2000-01-06 0.7\n",
"2000-01-07 0.7\n",
"dtype: float64"
]
}
],
"prompt_number": 37
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"Let us inspect the performance of the strategy."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"data[['EUROSTOXX', 'Dynamic']].plot(figsize=(10, 5))"
],
"language": "python",
"metadata": {
"slideshow": {
"slide_type": "-"
}
},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 38,
"text": [
"| \n", " | EUROSTOXX | \n", "VSTOXX | \n", "Equity | \n", "Volatility | \n", "Static | \n", "Dynamic | \n", "
|---|---|---|---|---|---|---|
| date | \n", "\n", " | \n", " | \n", " | \n", " | \n", " | \n", " |
| 2000-01-03 | \n", "100.000000 | \n", "100.000000 | \n", "0.700000 | \n", "0.300000 | \n", "100.000000 | \n", "100.000000 | \n", "
| 2000-01-04 | \n", "96.053180 | \n", "107.222966 | \n", "0.724420 | \n", "0.278124 | \n", "99.404116 | \n", "99.404116 | \n", "
| 2000-01-05 | \n", "93.659393 | \n", "105.195824 | \n", "0.725761 | \n", "0.276930 | \n", "97.120322 | \n", "97.106211 | \n", "
| 2000-01-06 | \n", "92.812659 | \n", "100.634511 | \n", "0.718221 | \n", "0.283884 | \n", "95.159214 | \n", "95.228521 | \n", "
| 2000-01-07 | \n", "95.856035 | \n", "88.562668 | \n", "0.686354 | \n", "0.318376 | \n", "93.668025 | \n", "93.987330 | \n", "
5 rows \u00d7 6 columns
\n", "\n",
"\n",
" \n",
"
\n",
"
"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 44,
"text": [
" bid bid_depth bid_depth_total offer \\\n",
"time \n",
"2014-03-12 10:00:00 534.850000 100.000000 100.000000 536.086000 \n",
"2014-03-12 10:05:00 534.850000 100.000000 100.000000 536.086000 \n",
"2014-03-12 10:10:00 535.355000 100.000000 100.000000 536.090000 \n",
"2014-03-12 10:15:00 535.054286 100.000000 100.000000 536.090000 \n",
"2014-03-12 10:20:00 534.600000 133.333333 133.333333 536.023333 \n",
"\n",
" offer_depth offer_depth_total \n",
"time \n",
"2014-03-12 10:00:00 100.000000 100.000000 \n",
"2014-03-12 10:05:00 100.000000 100.000000 \n",
"2014-03-12 10:10:00 100.000000 100.000000 \n",
"2014-03-12 10:15:00 142.857143 142.857143 \n",
"2014-03-12 10:20:00 116.666667 116.666667 \n",
"\n",
"[5 rows x 6 columns]"
]
}
],
"prompt_number": 44
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"Let's have a graphical look at the new data set."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"AAPL_5min['bid'].plot()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 45,
"text": [
"| \n", " | bid | \n", "bid_depth | \n", "bid_depth_total | \n", "offer | \n", "offer_depth | \n", "offer_depth_total | \n", "
|---|---|---|---|---|---|---|
| time | \n", "\n", " | \n", " | \n", " | \n", " | \n", " | \n", " |
| 2014-03-12 10:00:00 | \n", "534.850000 | \n", "100.000000 | \n", "100.000000 | \n", "536.086000 | \n", "100.000000 | \n", "100.000000 | \n", "
| 2014-03-12 10:05:00 | \n", "534.850000 | \n", "100.000000 | \n", "100.000000 | \n", "536.086000 | \n", "100.000000 | \n", "100.000000 | \n", "
| 2014-03-12 10:10:00 | \n", "535.355000 | \n", "100.000000 | \n", "100.000000 | \n", "536.090000 | \n", "100.000000 | \n", "100.000000 | \n", "
| 2014-03-12 10:15:00 | \n", "535.054286 | \n", "100.000000 | \n", "100.000000 | \n", "536.090000 | \n", "142.857143 | \n", "142.857143 | \n", "
| 2014-03-12 10:20:00 | \n", "534.600000 | \n", "133.333333 | \n", "133.333333 | \n", "536.023333 | \n", "116.666667 | \n", "116.666667 | \n", "
5 rows \u00d7 6 columns
\n", "![\"Continuum](\"CA_logo.png\"){kind=logo}
" ] }, { "cell_type": "heading", "level": 2, "metadata": { "slideshow": { "slide_type": "-" } }, "source": [ "Continuum Analytics Europe GmbH – Python Data Exploration & Visualization" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Continuum Analytics Inc.** – the company Web site\n", "\n", "www.continuum.io\n", "\n", "**Dr. Yves J. Hilpisch** – my personal Web site\n", "\n", "www.hilpisch.com\n", "\n", "**Python for Finance** – my NEW book (out as Early Release)\n", "\n", "Python for Finance (O'Reilly)\n", "\n", "**Derivatives Analytics with Python** – my current book\n", "\n", "www.derivatives-analytics-with-python.com\n", " \n", "**Contact Us**\n", "\n", "yves@continuum.io | europe@continuum.io\n", "| @dyjh | @ContinuumIO\n" ] } ], "metadata": {} } ] }