{
"cells": [
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"### The `pandas_datareader` Module\n",
"\n",
"In previous parts of this chapter, we discussed APIs. We use APIs to interface with data on a web server and bring that data in to Python.\n",
"\n",
"In this section, we make use of the `pandas_datareader` module. This is a module that provides a set of API features to Python in such a way that the data is directly loaded in to a `pandas` DataFrame format. The `pandas_datareader` module actually used to be part of the `pandas` module, but it was spun off in to its own separate module."
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"collapsed": false
},
"outputs": [
],
"source": [
"import pandas_datareader.data as web"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"Let's start with an example provided in the module's [documentation](https://pandas-datareader.readthedocs.io/en/latest/remote_data.html#fred)."
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
GDP
\n",
"
\n",
"
\n",
"
DATE
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
2005-01-01
\n",
"
12767.286
\n",
"
\n",
"
\n",
"
2005-04-01
\n",
"
12922.656
\n",
"
\n",
"
\n",
"
2005-07-01
\n",
"
13142.642
\n",
"
\n",
"
\n",
"
2005-10-01
\n",
"
13324.204
\n",
"
\n",
"
\n",
"
2006-01-01
\n",
"
13599.160
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" GDP\n",
"DATE \n",
"2005-01-01 12767.286\n",
"2005-04-01 12922.656\n",
"2005-07-01 13142.642\n",
"2005-10-01 13324.204\n",
"2006-01-01 13599.160"
]
},
"execution_count": 7,
"metadata": {
},
"output_type": "execute_result"
}
],
"source": [
"from datetime import datetime\n",
"\n",
"start = datetime(2005, 1, 1)\n",
"end = datetime(2019, 12, 31)\n",
"\n",
"gdp = web.DataReader('GDP', 'fred', start, end)\n",
"\n",
"gdp.head()"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"The above yields a data series of GDP values, listed at a quarterly frequency. Note that the `DATE` listed in the above DataFrame is not a column of data, it contains the labels for the rows. Remember, label names are bold-faced, so the name `2010-01-01` being in bold tells you that this is a row label, just like the name `GDP` in bold tells you that this is a column label.\n",
"\n",
"Above, we accessed data from the `'fred'` data provider. The FRED (Federal Reserve Economic Data) is an important provider of macroeconomic data, like measurement of gross domestic product.\n",
"\n",
"Note that we can collect multiple data series from FRED simultaneously by requesting a list of data series in the first argument to the `DataReader()` function, as shown below."
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
CPIAUCSL
\n",
"
CPILFESL
\n",
"
\n",
"
\n",
"
DATE
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
2005-01-01
\n",
"
191.6
\n",
"
199.0
\n",
"
\n",
"
\n",
"
2005-02-01
\n",
"
192.4
\n",
"
199.4
\n",
"
\n",
"
\n",
"
2005-03-01
\n",
"
193.1
\n",
"
200.1
\n",
"
\n",
"
\n",
"
2005-04-01
\n",
"
193.7
\n",
"
200.2
\n",
"
\n",
"
\n",
"
2005-05-01
\n",
"
193.6
\n",
"
200.5
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" CPIAUCSL CPILFESL\n",
"DATE \n",
"2005-01-01 191.6 199.0\n",
"2005-02-01 192.4 199.4\n",
"2005-03-01 193.1 200.1\n",
"2005-04-01 193.7 200.2\n",
"2005-05-01 193.6 200.5"
]
},
"execution_count": 8,
"metadata": {
},
"output_type": "execute_result"
}
],
"source": [
"inflation = web.DataReader(['CPIAUCSL', 'CPILFESL'], 'fred', start, end)\n",
"\n",
"inflation.head()"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"We see two columns of data printed above, corresponding to the two data series names that were requested. But how are these series names chosen? The name `'CPIAUCSL'`, for instance, is not an intuitive name for \"inflation.\" And if `'CPIAUCSL'` and `'CPILFESL'` are both data about inflation, what makes them different?\n",
"\n",
"If you search for a data series name on the [FRED website](https://fred.stlouisfed.org/), it will take you to the data series. Thus, we can easily find that:\n",
"- [CPIAUCSL](https://fred.stlouisfed.org/series/CPIAUCSL) is the consumer price index for urban consumers\n",
"- [CPILFESL](https://fred.stlouisfed.org/series/CPILFESL) is the consumer price index for urban consumers, excluding food and energy costs\n",
"\n",
"Data on FRED are organized by [category](https://fred.stlouisfed.org/categories),and [consumer price indexes](https://fred.stlouisfed.org/categories/9) are one such category.\n",
"\n",
"We could get the consumer price index for just alcohol using the data series `'CUSR0000SAF116'`."
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"
"
],
"text/plain": [
" DGS1MO DGS3MO DGS6MO DGS1 DGS2 DGS5 DGS7 DGS10 DGS30\n",
"DATE \n",
"2005-01-03 1.99 2.32 2.63 2.79 3.10 3.64 3.94 4.23 4.85\n",
"2005-01-04 2.05 2.33 2.63 2.82 3.20 3.72 4.02 4.29 4.91\n",
"2005-01-05 2.04 2.33 2.63 2.83 3.22 3.73 4.02 4.29 4.88\n",
"2005-01-06 2.04 2.31 2.63 2.82 3.18 3.71 4.01 4.29 4.89\n",
"2005-01-07 2.03 2.32 2.63 2.82 3.20 3.73 4.03 4.29 4.88"
]
},
"execution_count": 10,
"metadata": {
},
"output_type": "execute_result"
}
],
"source": [
"data_list = ['DGS1MO', 'DGS3MO', 'DGS6MO', 'DGS1', 'DGS2', 'DGS5', 'DGS7', 'DGS10', 'DGS30']\n",
"\n",
"ts = web.DataReader(data_list, 'fred', start, end)\n",
"\n",
"ts.head()"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"The data returned provides the term structure of interest rates. Interest rates on treasuries of varying maturities are collected.\n",
"\n",
"It's time to treat these row labels a bit more formally.\n",
"\n",
"Recall that column labels can be accessed like so."
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"Index(['DGS1MO', 'DGS3MO', 'DGS6MO', 'DGS1', 'DGS2', 'DGS5', 'DGS7', 'DGS10',\n",
" 'DGS30'],\n",
" dtype='object')"
]
},
"execution_count": 11,
"metadata": {
},
"output_type": "execute_result"
}
],
"source": [
"ts.columns"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"The columns of a DataFrame are referred to as, quite intuitively, *columns*!\n",
"\n",
"The row labels are a bit different, they are typically referred to as the *index* of the DataFrame. The list of row labels is thus accessible via:"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"DatetimeIndex(['2005-01-03', '2005-01-04', '2005-01-05', '2005-01-06',\n",
" '2005-01-07', '2005-01-10', '2005-01-11', '2005-01-12',\n",
" '2005-01-13', '2005-01-14',\n",
" ...\n",
" '2019-12-18', '2019-12-19', '2019-12-20', '2019-12-23',\n",
" '2019-12-24', '2019-12-25', '2019-12-26', '2019-12-27',\n",
" '2019-12-30', '2019-12-31'],\n",
" dtype='datetime64[ns]', name='DATE', length=3912, freq=None)"
]
},
"execution_count": 12,
"metadata": {
},
"output_type": "execute_result"
}
],
"source": [
"ts.index"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"It is at this point that use of `.loc` becomes interesting and useful, rather than simply an academic point about the structure of DataFrames. For instance, to look up the term structure of interest rates on June 7, 2019, we would enter:"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"DGS1MO 2.30\n",
"DGS3MO 2.28\n",
"DGS6MO 2.15\n",
"DGS1 1.97\n",
"DGS2 1.85\n",
"DGS5 1.85\n",
"DGS7 1.97\n",
"DGS10 2.09\n",
"DGS30 2.57\n",
"Name: 2019-06-07 00:00:00, dtype: float64"
]
},
"execution_count": 13,
"metadata": {
},
"output_type": "execute_result"
}
],
"source": [
"ts.loc['2019-06-07']"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"It's helpful to have an index because it lets us label the data in two dimensions at once. The column label will tell us the name of the data series (e.g. `'DGS1'`, which is the yield on a set of 1 year constant maturity U.S. treasuries) and the row *index* will tell us the time at which that data is observed.\n",
"\n",
"In earlier parts of this book, the row index was boring. Unless explicitly given a list of row labels to use for the index, Python will assign the row index to be the row number. For example:"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
x
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
1.764052
\n",
"
\n",
"
\n",
"
1
\n",
"
0.400157
\n",
"
\n",
"
\n",
"
2
\n",
"
0.978738
\n",
"
\n",
"
\n",
"
3
\n",
"
2.240893
\n",
"
\n",
"
\n",
"
4
\n",
"
1.867558
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" x\n",
"0 1.764052\n",
"1 0.400157\n",
"2 0.978738\n",
"3 2.240893\n",
"4 1.867558"
]
},
"execution_count": 14,
"metadata": {
},
"output_type": "execute_result"
}
],
"source": [
"import pandas as pd\n",
"import numpy as np\n",
"np.random.seed(0)\n",
"df = pd.DataFrame(columns=['x'])\n",
"df['x'] = np.random.normal(0, 1, 10)\n",
"df.head()"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"Note that the data returned from `pandas_datareader` is indexed in a very convenient way. Along with using `ts.loc['2019-06-07']`, we can get slices of the data like so:"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
DGS1MO
\n",
"
DGS3MO
\n",
"
DGS6MO
\n",
"
DGS1
\n",
"
DGS2
\n",
"
DGS5
\n",
"
DGS7
\n",
"
DGS10
\n",
"
DGS30
\n",
"
\n",
"
\n",
"
DATE
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
2005-01-03
\n",
"
1.99
\n",
"
2.32
\n",
"
2.63
\n",
"
2.79
\n",
"
3.10
\n",
"
3.64
\n",
"
3.94
\n",
"
4.23
\n",
"
4.85
\n",
"
\n",
"
\n",
"
2005-01-04
\n",
"
2.05
\n",
"
2.33
\n",
"
2.63
\n",
"
2.82
\n",
"
3.20
\n",
"
3.72
\n",
"
4.02
\n",
"
4.29
\n",
"
4.91
\n",
"
\n",
"
\n",
"
2005-01-05
\n",
"
2.04
\n",
"
2.33
\n",
"
2.63
\n",
"
2.83
\n",
"
3.22
\n",
"
3.73
\n",
"
4.02
\n",
"
4.29
\n",
"
4.88
\n",
"
\n",
"
\n",
"
2005-01-06
\n",
"
2.04
\n",
"
2.31
\n",
"
2.63
\n",
"
2.82
\n",
"
3.18
\n",
"
3.71
\n",
"
4.01
\n",
"
4.29
\n",
"
4.89
\n",
"
\n",
"
\n",
"
2005-01-07
\n",
"
2.03
\n",
"
2.32
\n",
"
2.63
\n",
"
2.82
\n",
"
3.20
\n",
"
3.73
\n",
"
4.03
\n",
"
4.29
\n",
"
4.88
\n",
"
\n",
"
\n",
"
2005-01-10
\n",
"
2.07
\n",
"
2.36
\n",
"
2.67
\n",
"
2.86
\n",
"
3.23
\n",
"
3.75
\n",
"
4.03
\n",
"
4.29
\n",
"
4.86
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" DGS1MO DGS3MO DGS6MO DGS1 DGS2 DGS5 DGS7 DGS10 DGS30\n",
"DATE \n",
"2005-01-03 1.99 2.32 2.63 2.79 3.10 3.64 3.94 4.23 4.85\n",
"2005-01-04 2.05 2.33 2.63 2.82 3.20 3.72 4.02 4.29 4.91\n",
"2005-01-05 2.04 2.33 2.63 2.83 3.22 3.73 4.02 4.29 4.88\n",
"2005-01-06 2.04 2.31 2.63 2.82 3.18 3.71 4.01 4.29 4.89\n",
"2005-01-07 2.03 2.32 2.63 2.82 3.20 3.73 4.03 4.29 4.88\n",
"2005-01-10 2.07 2.36 2.67 2.86 3.23 3.75 4.03 4.29 4.86"
]
},
"execution_count": 15,
"metadata": {
},
"output_type": "execute_result"
}
],
"source": [
"ts.loc['2005-01-03':'2005-01-10']"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"There are two important things to notice about index slicing here (*slicing* is another term for getting a subset of data).\n",
"\n",
"First, the slice ends *at* the endpoint, not just before it. *This breaks from normal Python convention.* For instance, `range(0:10)` will give you the numbers 0 through 9, not the numbers 0 through 10. When slicing by index, the last row label you include is the last row you get. As another example:"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
x
\n",
"
\n",
" \n",
" \n",
"
\n",
"
1
\n",
"
0.400157
\n",
"
\n",
"
\n",
"
2
\n",
"
0.978738
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" x\n",
"1 0.400157\n",
"2 0.978738"
]
},
"execution_count": 16,
"metadata": {
},
"output_type": "execute_result"
}
],
"source": [
"df.loc[1:2]"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"Second, the data exists for Jan. 3-7 and Jan. 10, but not for Jan. 8-9. Why? Because Jan. 8-9 was a weekend.\n",
"\n",
"We can shift date indexes around quite easily via `.shift()`. This will come in handy in many applications."
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"
"
],
"text/plain": [
" DGS1MO DGS3MO DGS6MO DGS1 DGS2 DGS5 DGS7 DGS10 DGS30\n",
"DATE \n",
"2005-01-03 NaN NaN NaN NaN NaN NaN NaN NaN NaN\n",
"2005-01-04 1.99 2.32 2.63 2.79 3.10 3.64 3.94 4.23 4.85\n",
"2005-01-05 2.05 2.33 2.63 2.82 3.20 3.72 4.02 4.29 4.91\n",
"2005-01-06 2.04 2.33 2.63 2.83 3.22 3.73 4.02 4.29 4.88\n",
"2005-01-07 2.04 2.31 2.63 2.82 3.18 3.71 4.01 4.29 4.89\n",
"2005-01-10 2.03 2.32 2.63 2.82 3.20 3.73 4.03 4.29 4.88"
]
},
"execution_count": 18,
"metadata": {
},
"output_type": "execute_result"
}
],
"source": [
"ts.shift(1).head(6)"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"Note that the shifting moves data based on the dates included in the dataset. For instance, the interest rates on Friday, if we shift the data forward by one day, are moved to Monday and not to Saturday.\n",
"\n",
"Shifting is incredibly useful for calculating things like growth rates."
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"
"
]
},
"execution_count": 24,
"metadata": {
"image/png": {
"height": 411,
"width": 720
},
"needs_background": "light"
},
"output_type": "execute_result"
}
],
"source": [
"import seaborn as sns\n",
"sns.lineplot(x=mat, y=ts.loc['2016-06-07', data_list])"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"The yield curve does not always have this shape to it, though what's shown above is the \"standard\" appearance. Typically, the yield curve is upward-sloping and somewhat curvy.\n",
"\n",
"In contrast, the yield curve may sometimes invert. That is, it will have places in the curve that are downward-sloping."
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 25,
"metadata": {
},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"execution_count": 25,
"metadata": {
"image/png": {
"height": 411,
"width": 720
},
"needs_background": "light"
},
"output_type": "execute_result"
}
],
"source": [
"sns.lineplot(x=mat, y=ts.loc['2007-01-02', data_list])"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"The usual place where we look for this is in the 10-year Treasury minus the 2-year Treasury."
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 27,
"metadata": {
},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"execution_count": 27,
"metadata": {
"image/png": {
"height": 424,
"width": 706
},
"needs_background": "light"
},
"output_type": "execute_result"
}
],
"source": [
"ts['Slope'] = ts['DGS10'] - ts['DGS2']\n",
"ts['Slope'].plot()"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"When this slope (the difference between the 10 year and 2 year) is negative, we refer to this as a yield curve inversion.\n",
"\n",
"These inversion events are reasonably good predictors of recessions.\n",
"\n",
"Let's test this idea. Begin by getting a series of data about when recessions occur from [FRED](https://fred.stlouisfed.org/series/JHDUSRGDPBR)."
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
JHDUSRGDPBR
\n",
"
\n",
"
\n",
"
DATE
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
2005-01-01
\n",
"
0.0
\n",
"
\n",
"
\n",
"
2005-04-01
\n",
"
0.0
\n",
"
\n",
"
\n",
"
2005-07-01
\n",
"
0.0
\n",
"
\n",
"
\n",
"
2005-10-01
\n",
"
0.0
\n",
"
\n",
"
\n",
"
2006-01-01
\n",
"
0.0
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" JHDUSRGDPBR\n",
"DATE \n",
"2005-01-01 0.0\n",
"2005-04-01 0.0\n",
"2005-07-01 0.0\n",
"2005-10-01 0.0\n",
"2006-01-01 0.0"
]
},
"execution_count": 28,
"metadata": {
},
"output_type": "execute_result"
}
],
"source": [
"recession = web.DataReader('JHDUSRGDPBR', 'fred', start, end)\n",
"recession.head()"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"The data series `'JHDUSRGDPBR'` is a list of recessions. The value will be 1 when there is a recession and 0 when there is not a recession.\n",
"\n",
"Let's now merge the recession data. Note that the recession data is observed once per quarter. We should convert all data to a quarterly frequency before preceeding."
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"
"
],
"text/plain": [
" JHDUSRGDPBR\n",
"DATE \n",
"2005Q1 0.0\n",
"2005Q2 0.0\n",
"2005Q3 0.0\n",
"2005Q4 0.0\n",
"2006Q1 0.0"
]
},
"execution_count": 30,
"metadata": {
},
"output_type": "execute_result"
}
],
"source": [
"recession = recession.to_period('Q')\n",
"recession.head()"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"Note that the term structure data has multiple observations for a given quarter. Suppose that we only want to use the first observation. That is, at the start of a quarter, what does the term structure look like? We can remove duplicate index values to get rid of the observations that do not correspond to the first observation for each quarter using `.duplicated()`."
]
},
{
"cell_type": "code",
"execution_count": 31,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
DGS1MO
\n",
"
DGS3MO
\n",
"
DGS6MO
\n",
"
DGS1
\n",
"
DGS2
\n",
"
DGS5
\n",
"
DGS7
\n",
"
DGS10
\n",
"
DGS30
\n",
"
DGS30 growth
\n",
"
Slope
\n",
"
\n",
"
\n",
"
DATE
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
2005Q1
\n",
"
1.99
\n",
"
2.32
\n",
"
2.63
\n",
"
2.79
\n",
"
3.10
\n",
"
3.64
\n",
"
3.94
\n",
"
4.23
\n",
"
4.85
\n",
"
NaN
\n",
"
1.13
\n",
"
\n",
"
\n",
"
2005Q2
\n",
"
2.66
\n",
"
2.80
\n",
"
3.13
\n",
"
3.34
\n",
"
3.75
\n",
"
4.13
\n",
"
4.29
\n",
"
4.46
\n",
"
4.72
\n",
"
-0.008403
\n",
"
0.71
\n",
"
\n",
"
\n",
"
2005Q3
\n",
"
3.02
\n",
"
3.17
\n",
"
3.38
\n",
"
3.51
\n",
"
3.76
\n",
"
3.84
\n",
"
3.92
\n",
"
4.06
\n",
"
4.28
\n",
"
0.021480
\n",
"
0.30
\n",
"
\n",
"
\n",
"
2005Q4
\n",
"
3.22
\n",
"
3.61
\n",
"
4.02
\n",
"
4.09
\n",
"
4.21
\n",
"
4.25
\n",
"
4.31
\n",
"
4.39
\n",
"
4.58
\n",
"
0.011038
\n",
"
0.18
\n",
"
\n",
"
\n",
"
2006Q1
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" DGS1MO DGS3MO DGS6MO DGS1 DGS2 DGS5 DGS7 DGS10 DGS30 \\\n",
"DATE \n",
"2005Q1 1.99 2.32 2.63 2.79 3.10 3.64 3.94 4.23 4.85 \n",
"2005Q2 2.66 2.80 3.13 3.34 3.75 4.13 4.29 4.46 4.72 \n",
"2005Q3 3.02 3.17 3.38 3.51 3.76 3.84 3.92 4.06 4.28 \n",
"2005Q4 3.22 3.61 4.02 4.09 4.21 4.25 4.31 4.39 4.58 \n",
"2006Q1 NaN NaN NaN NaN NaN NaN NaN NaN NaN \n",
"\n",
" DGS30 growth Slope \n",
"DATE \n",
"2005Q1 NaN 1.13 \n",
"2005Q2 -0.008403 0.71 \n",
"2005Q3 0.021480 0.30 \n",
"2005Q4 0.011038 0.18 \n",
"2006Q1 NaN NaN "
]
},
"execution_count": 31,
"metadata": {
},
"output_type": "execute_result"
}
],
"source": [
"tsq = tsq[~tsq.index.duplicated(keep='first')]\n",
"tsq.head()"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"The observation at `'2006Q1'` is troubling. Why is there missing data there?"
]
},
{
"cell_type": "code",
"execution_count": 32,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
DGS1MO
\n",
"
DGS3MO
\n",
"
DGS6MO
\n",
"
DGS1
\n",
"
DGS2
\n",
"
DGS5
\n",
"
DGS7
\n",
"
DGS10
\n",
"
DGS30
\n",
"
DGS30 growth
\n",
"
Slope
\n",
"
\n",
"
\n",
"
DATE
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
2005-12-29
\n",
"
3.67
\n",
"
4.01
\n",
"
4.34
\n",
"
4.35
\n",
"
4.38
\n",
"
4.33
\n",
"
4.34
\n",
"
4.37
\n",
"
4.49
\n",
"
-0.004435
\n",
"
-0.01
\n",
"
\n",
"
\n",
"
2005-12-30
\n",
"
4.01
\n",
"
4.08
\n",
"
4.37
\n",
"
4.38
\n",
"
4.41
\n",
"
4.35
\n",
"
4.36
\n",
"
4.39
\n",
"
4.51
\n",
"
0.004454
\n",
"
-0.02
\n",
"
\n",
"
\n",
"
2006-01-02
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
NaN
\n",
"
\n",
"
\n",
"
2006-01-03
\n",
"
4.05
\n",
"
4.16
\n",
"
4.40
\n",
"
4.38
\n",
"
4.34
\n",
"
4.30
\n",
"
4.32
\n",
"
4.37
\n",
"
4.52
\n",
"
NaN
\n",
"
0.03
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" DGS1MO DGS3MO DGS6MO DGS1 DGS2 DGS5 DGS7 DGS10 DGS30 \\\n",
"DATE \n",
"2005-12-29 3.67 4.01 4.34 4.35 4.38 4.33 4.34 4.37 4.49 \n",
"2005-12-30 4.01 4.08 4.37 4.38 4.41 4.35 4.36 4.39 4.51 \n",
"2006-01-02 NaN NaN NaN NaN NaN NaN NaN NaN NaN \n",
"2006-01-03 4.05 4.16 4.40 4.38 4.34 4.30 4.32 4.37 4.52 \n",
"\n",
" DGS30 growth Slope \n",
"DATE \n",
"2005-12-29 -0.004435 -0.01 \n",
"2005-12-30 0.004454 -0.02 \n",
"2006-01-02 NaN NaN \n",
"2006-01-03 NaN 0.03 "
]
},
"execution_count": 32,
"metadata": {
},
"output_type": "execute_result"
}
],
"source": [
"ts.loc['2005-12-29':'2006-01-03']"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"It's unclear why the first observation for the first quarter of 2006 is missing, but we don't want to lose that quarter of data simply because one date has missing data. Let's fill in the data. The method we'll use to do so is `'ffill'`, which stands for forward fill. Pandas will take the most recent data (in this case `'2005-12-30'`) and use that to fill in the missing data (at `'2006-01-02'`)."
]
},
{
"cell_type": "code",
"execution_count": 33,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"
"
],
"text/plain": [
" DGS1MO DGS3MO DGS6MO DGS1 DGS2 DGS5 DGS7 DGS10 DGS30 \\\n",
"DATE \n",
"2005Q1 1.99 2.32 2.63 2.79 3.10 3.64 3.94 4.23 4.85 \n",
"2005Q2 2.66 2.80 3.13 3.34 3.75 4.13 4.29 4.46 4.72 \n",
"2005Q3 3.02 3.17 3.38 3.51 3.76 3.84 3.92 4.06 4.28 \n",
"2005Q4 3.22 3.61 4.02 4.09 4.21 4.25 4.31 4.39 4.58 \n",
"2006Q1 4.01 4.08 4.37 4.38 4.41 4.35 4.36 4.39 4.51 \n",
"\n",
" DGS30 growth Slope JHDUSRGDPBR \n",
"DATE \n",
"2005Q1 NaN 1.13 0.0 \n",
"2005Q2 -0.008403 0.71 0.0 \n",
"2005Q3 0.021480 0.30 0.0 \n",
"2005Q4 0.011038 0.18 0.0 \n",
"2006Q1 0.004454 -0.02 0.0 "
]
},
"execution_count": 36,
"metadata": {
},
"output_type": "execute_result"
}
],
"source": [
"ts_recession = tsq.merge(recession, on='DATE')\n",
"ts_recession.head()"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"And finally, we can try to see if recessions are more likely when the slope is negative. In order to say that yield curve inversions might predict recessions, we need to use data from before a recession begins. For instance, does an inversion at quarter 3 of year $t$ imply that a recession is likely to begin at quarter 4 of year $t$? We'll use `.shift()` to make a lagged copy of the `'Slope'` column."
]
},
{
"cell_type": "code",
"execution_count": 38,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
DGS1MO
\n",
"
DGS3MO
\n",
"
DGS6MO
\n",
"
DGS1
\n",
"
DGS2
\n",
"
DGS5
\n",
"
DGS7
\n",
"
DGS10
\n",
"
DGS30
\n",
"
DGS30 growth
\n",
"
Slope
\n",
"
JHDUSRGDPBR
\n",
"
Lagged Slope
\n",
"
\n",
"
\n",
"
DATE
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
2005Q1
\n",
"
1.99
\n",
"
2.32
\n",
"
2.63
\n",
"
2.79
\n",
"
3.10
\n",
"
3.64
\n",
"
3.94
\n",
"
4.23
\n",
"
4.85
\n",
"
NaN
\n",
"
1.13
\n",
"
0.0
\n",
"
NaN
\n",
"
\n",
"
\n",
"
2005Q2
\n",
"
2.66
\n",
"
2.80
\n",
"
3.13
\n",
"
3.34
\n",
"
3.75
\n",
"
4.13
\n",
"
4.29
\n",
"
4.46
\n",
"
4.72
\n",
"
-0.008403
\n",
"
0.71
\n",
"
0.0
\n",
"
1.13
\n",
"
\n",
"
\n",
"
2005Q3
\n",
"
3.02
\n",
"
3.17
\n",
"
3.38
\n",
"
3.51
\n",
"
3.76
\n",
"
3.84
\n",
"
3.92
\n",
"
4.06
\n",
"
4.28
\n",
"
0.021480
\n",
"
0.30
\n",
"
0.0
\n",
"
0.71
\n",
"
\n",
"
\n",
"
2005Q4
\n",
"
3.22
\n",
"
3.61
\n",
"
4.02
\n",
"
4.09
\n",
"
4.21
\n",
"
4.25
\n",
"
4.31
\n",
"
4.39
\n",
"
4.58
\n",
"
0.011038
\n",
"
0.18
\n",
"
0.0
\n",
"
0.30
\n",
"
\n",
"
\n",
"
2006Q1
\n",
"
4.01
\n",
"
4.08
\n",
"
4.37
\n",
"
4.38
\n",
"
4.41
\n",
"
4.35
\n",
"
4.36
\n",
"
4.39
\n",
"
4.51
\n",
"
0.004454
\n",
"
-0.02
\n",
"
0.0
\n",
"
0.18
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" DGS1MO DGS3MO DGS6MO DGS1 DGS2 DGS5 DGS7 DGS10 DGS30 \\\n",
"DATE \n",
"2005Q1 1.99 2.32 2.63 2.79 3.10 3.64 3.94 4.23 4.85 \n",
"2005Q2 2.66 2.80 3.13 3.34 3.75 4.13 4.29 4.46 4.72 \n",
"2005Q3 3.02 3.17 3.38 3.51 3.76 3.84 3.92 4.06 4.28 \n",
"2005Q4 3.22 3.61 4.02 4.09 4.21 4.25 4.31 4.39 4.58 \n",
"2006Q1 4.01 4.08 4.37 4.38 4.41 4.35 4.36 4.39 4.51 \n",
"\n",
" DGS30 growth Slope JHDUSRGDPBR Lagged Slope \n",
"DATE \n",
"2005Q1 NaN 1.13 0.0 NaN \n",
"2005Q2 -0.008403 0.71 0.0 1.13 \n",
"2005Q3 0.021480 0.30 0.0 0.71 \n",
"2005Q4 0.011038 0.18 0.0 0.30 \n",
"2006Q1 0.004454 -0.02 0.0 0.18 "
]
},
"execution_count": 38,
"metadata": {
},
"output_type": "execute_result"
}
],
"source": [
"ts_recession['Lagged Slope'] = ts_recession['Slope'].shift(1)\n",
"ts_recession.head()"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"One last data-cleaning issue: `statsmodels` *hates* missing values. It throws a fit and stops working when it sees them. Notice that the first value of `'Lagged Slope'` is missing, because we can't observe a lag for the first data point.\n",
"\n",
"To get rid of it, we use `.dropna()` to find all observations where the data is not missing."
]
},
{
"cell_type": "code",
"execution_count": 39,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
DGS1MO
\n",
"
DGS3MO
\n",
"
DGS6MO
\n",
"
DGS1
\n",
"
DGS2
\n",
"
DGS5
\n",
"
DGS7
\n",
"
DGS10
\n",
"
DGS30
\n",
"
DGS30 growth
\n",
"
Slope
\n",
"
JHDUSRGDPBR
\n",
"
Lagged Slope
\n",
"
\n",
"
\n",
"
DATE
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
2005Q2
\n",
"
2.66
\n",
"
2.80
\n",
"
3.13
\n",
"
3.34
\n",
"
3.75
\n",
"
4.13
\n",
"
4.29
\n",
"
4.46
\n",
"
4.72
\n",
"
-0.008403
\n",
"
0.71
\n",
"
0.0
\n",
"
1.13
\n",
"
\n",
"
\n",
"
2005Q3
\n",
"
3.02
\n",
"
3.17
\n",
"
3.38
\n",
"
3.51
\n",
"
3.76
\n",
"
3.84
\n",
"
3.92
\n",
"
4.06
\n",
"
4.28
\n",
"
0.021480
\n",
"
0.30
\n",
"
0.0
\n",
"
0.71
\n",
"
\n",
"
\n",
"
2005Q4
\n",
"
3.22
\n",
"
3.61
\n",
"
4.02
\n",
"
4.09
\n",
"
4.21
\n",
"
4.25
\n",
"
4.31
\n",
"
4.39
\n",
"
4.58
\n",
"
0.011038
\n",
"
0.18
\n",
"
0.0
\n",
"
0.30
\n",
"
\n",
"
\n",
"
2006Q1
\n",
"
4.01
\n",
"
4.08
\n",
"
4.37
\n",
"
4.38
\n",
"
4.41
\n",
"
4.35
\n",
"
4.36
\n",
"
4.39
\n",
"
4.51
\n",
"
0.004454
\n",
"
-0.02
\n",
"
0.0
\n",
"
0.18
\n",
"
\n",
"
\n",
"
2006Q2
\n",
"
4.66
\n",
"
4.67
\n",
"
4.86
\n",
"
4.86
\n",
"
4.86
\n",
"
4.85
\n",
"
4.86
\n",
"
4.88
\n",
"
4.90
\n",
"
0.000000
\n",
"
0.02
\n",
"
0.0
\n",
"
-0.02
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" DGS1MO DGS3MO DGS6MO DGS1 DGS2 DGS5 DGS7 DGS10 DGS30 \\\n",
"DATE \n",
"2005Q2 2.66 2.80 3.13 3.34 3.75 4.13 4.29 4.46 4.72 \n",
"2005Q3 3.02 3.17 3.38 3.51 3.76 3.84 3.92 4.06 4.28 \n",
"2005Q4 3.22 3.61 4.02 4.09 4.21 4.25 4.31 4.39 4.58 \n",
"2006Q1 4.01 4.08 4.37 4.38 4.41 4.35 4.36 4.39 4.51 \n",
"2006Q2 4.66 4.67 4.86 4.86 4.86 4.85 4.86 4.88 4.90 \n",
"\n",
" DGS30 growth Slope JHDUSRGDPBR Lagged Slope \n",
"DATE \n",
"2005Q2 -0.008403 0.71 0.0 1.13 \n",
"2005Q3 0.021480 0.30 0.0 0.71 \n",
"2005Q4 0.011038 0.18 0.0 0.30 \n",
"2006Q1 0.004454 -0.02 0.0 0.18 \n",
"2006Q2 0.000000 0.02 0.0 -0.02 "
]
},
"execution_count": 39,
"metadata": {
},
"output_type": "execute_result"
}
],
"source": [
"ts_recession = ts_recession.dropna()\n",
"ts_recession.head()"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"Now, we can use `statsmodels` to run a logistic regression (since the y variable, `'JHDUSRGDPBR'` is only ever 0 or 1).\n",
"\n",
"One tricky element here is that the name for our $x$ variable is `'Lagged Slope'`. Python will get confused by this; if we write `'JHDUSRGDPBR ~ Lagged Slope'` then Python will parse this as a function\n",
"$$\n",
"\\text{JHDUSRGDPBR} = \\beta_0 + \\beta_1\\text{Lagged} + \\beta_2\\text{Slope} + u\n",
"$$\n",
"which is not what we want. To ensure that Python reads `'Lagged Slope'` as one column name, use the `Q()` function. That is, refer to the variable as `Q(\"Lagged Slope\")`."
]
},
{
"cell_type": "code",
"execution_count": 42,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Optimization terminated successfully.\n",
" Current function value: 0.363198\n",
" Iterations 6\n",
" Logit Regression Results \n",
"==============================================================================\n",
"Dep. Variable: JHDUSRGDPBR No. Observations: 59\n",
"Model: Logit Df Residuals: 57\n",
"Method: MLE Df Model: 1\n",
"Date: Sun, 26 Sep 2021 Pseudo R-squ.: 0.002790\n",
"Time: 21:01:12 Log-Likelihood: -21.429\n",
"converged: True LL-Null: -21.489\n",
"Covariance Type: nonrobust LLR p-value: 0.7291\n",
"=====================================================================================\n",
" coef std err z P>|z| [0.025 0.975]\n",
"-------------------------------------------------------------------------------------\n",
"Intercept -1.8067 0.685 -2.637 0.008 -3.150 -0.464\n",
"Q(\"Lagged Slope\") -0.1601 0.464 -0.345 0.730 -1.069 0.749\n",
"=====================================================================================\n"
]
}
],
"source": [
"import statsmodels.formula.api as smf\n",
"\n",
"model = smf.logit('JHDUSRGDPBR ~ Q(\"Lagged Slope\")', data=ts_recession).fit()\n",
"print(model.summary())"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"Negative $\\beta$ coefficients in a Logit model mean that the covariate (the right hand side variable) corresponding to that $\\beta$ coefficient are negatively predictive of the event. The \"event\" is an instance of y=1. So, what we see here is that a higher slope is negatively predictive of a recession.\n",
"\n",
"If you load a much longer dataset, you can see a similar pattern over time.\n",
"\n",
"The question is, why does this work? What's the connection between the yield curve and recessions?\n",
"\n",
"Let's turn our attention to the *expectations hypothesis*, which is covered in earlier courses. Define $y_t^{(i)}$ to be the one period interest rate at time $t$ for a Treasury security with $i$ periods to maturity. We thus have\n",
"$$\n",
"\\Big(1+y_t^{(2)}\\Big)^2 = \\Big(1+y_t^{(1)}\\Big)\\text{E}_t\\Big[\\Big(1+y_{t+1}^{(1)}\\Big)\\Big]\n",
"$$\n",
"\n",
"That is, the cumulative return for holding a two-year Treasury to maturity should be the same return as holding a one-year Treasury to maturity and then holding another (future) one-year Treasury to maturity after that. At year $t$, we don't know what the one-year Treasury will be at in year $t+1$. Thus, we say that the *expected* value of it is $\\text{E}_t\\Big[\\Big(1+y_{t+1}^{(1)}\\Big)\\Big]$, where $\\text{E}$ stands for expected value.\n",
"\n",
"Let's define the one-year forward rate as\n",
"$$\n",
"\\Big(1+f_{t}^{(1)}\\Big) = \\frac{\\Big(1+y_t^{(2)}\\Big)^2}{\\Big(1+y_t^{(1)}\\Big)}\n",
"$$\n",
"\n",
"The value for $f_{t}^{(1)}$ is something that we can calculate given other data at time $t$. The Expectations Hypothesis says that this value should be a good estimate for the expected future spot rate, i.e.:\n",
"$$\n",
"f_{t}^{(1)} = \\text{E}_t\\Big[1+y_{t+1}^{(1)}\\Big]\n",
"$$\n",
"\n",
"Compute the forward rate below."
]
},
{
"cell_type": "code",
"execution_count": 43,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"