notebooks/L03_pandas/v01_pandas_intro.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Introduction\n",
    "\n",
    "**Prerequisites**\n",
    "\n",
    "- [Python Fundamentals](../python_fundamentals/index.ipynb)  \n",
    "\n",
    "\n",
    "**Outcomes**\n",
    "\n",
    "- Understand the core pandas objects (`Series` and `DataFrame`)  \n",
    "- Index into particular elements of a Series and DataFrame  \n",
    "- Understand what `.dtype`/`.dtypes` do  \n",
    "- Make basic visualizations  \n",
    "\n",
    "\n",
    "**Data**\n",
    "\n",
    "- US regional unemployment data from Bureau of Labor Statistics  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Outline\n",
    "\n",
    "- [Introduction](#Introduction)  \n",
    "  - [pandas](#pandas)  \n",
    "  - [Series](#Series)  \n",
    "  - [DataFrame](#DataFrame)  \n",
    "  - [Data Types](#Data-Types)  \n",
    "  - [Changing DataFrames](#Changing-DataFrames)  \n",
    "  - [Exercises](#Exercises)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "hide-output": false
   },
   "outputs": [],
   "source": [
    "# Uncomment following line to install on colab\n",
    "#! pip install qeds"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## pandas\n",
    "\n",
    "This lecture begins the material on `pandas`.\n",
    "\n",
    "To start, we will import the pandas package and give it the alias\n",
    "`pd`, which is conventional practice."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "hide-output": false
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "\n",
    "# Don't worry about this line for now!\n",
    "%matplotlib inline\n",
    "# activate plot theme\n",
    "import qeds\n",
    "qeds.themes.mpl_style();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Sometimes, knowing which pandas version we are\n",
    "using is helpful.\n",
    "\n",
    "We can check this by running the code below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'0.24.2'"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.__version__"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Series\n",
    "\n",
    "The first main pandas type we will introduce is called Series.\n",
    "\n",
    "A Series is a single column of data, with row labels for each\n",
    "observation.\n",
    "\n",
    "pandas refers to the row labels as the *index* of the Series.\n",
    "\n",
    "<img src=\"https://datascience.quantecon.org/assets/_static/intro_files/PandasSeries.png\" alt=\"PandasSeries.png\" style=\"\">\n",
    "\n",
    "  \n",
    "Below, we create a Series which contains the US unemployment rate every\n",
    "other year starting in 1995."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "hide-output": false
   },
   "outputs": [],
   "source": [
    "values = [5.6, 5.3, 4.3, 4.2, 5.8, 5.3, 4.6, 7.8, 9.1, 8., 5.7]\n",
    "years = list(range(1995, 2017, 2))\n",
    "\n",
    "unemp = pd.Series(data=values, index=years, name=\"Unemployment\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1995    5.6\n",
       "1997    5.3\n",
       "1999    4.3\n",
       "2001    4.2\n",
       "2003    5.8\n",
       "2005    5.3\n",
       "2007    4.6\n",
       "2009    7.8\n",
       "2011    9.1\n",
       "2013    8.0\n",
       "2015    5.7\n",
       "Name: Unemployment, dtype: float64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can look at the index and values in our Series."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Int64Index([1995, 1997, 1999, 2001, 2003, 2005, 2007, 2009, 2011, 2013, 2015], dtype='int64')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([5.6, 5.3, 4.3, 4.2, 5.8, 5.3, 4.6, 7.8, 9.1, 8. , 5.7])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp.values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### What Can We Do with a Series object?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### `.head` and `.tail`\n",
    "\n",
    "Often, our data will have many rows, and we won’t want to display it all\n",
    "at once.\n",
    "\n",
    "The methods `.head` and `.tail` show rows at the beginning and end\n",
    "of our Series, respectively."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1995    5.6\n",
       "1997    5.3\n",
       "1999    4.3\n",
       "2001    4.2\n",
       "2003    5.8\n",
       "Name: Unemployment, dtype: float64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2007    4.6\n",
       "2009    7.8\n",
       "2011    9.1\n",
       "2013    8.0\n",
       "2015    5.7\n",
       "Name: Unemployment, dtype: float64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Basic Plotting\n",
    "\n",
    "We can also plot data using the `.plot` method."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7ff4e4364ef0>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "unemp.plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">**Note**\n",
    ">\n",
    ">This is why we needed the `%matplotlib inline` — it tells the notebook\n",
    "to display figures inside the notebook itself. Also, pandas has much greater visualization functionality than this, but we will study that later on."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Unique Values\n",
    "\n",
    "Though it doesn’t make sense in this data set, we may want to find the\n",
    "unique values in a Series – which can be done with the `.unique` method."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([5.6, 5.3, 4.3, 4.2, 5.8, 4.6, 7.8, 9.1, 8. , 5.7])"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp.unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Indexing\n",
    "\n",
    "Sometimes, we will want to select particular elements from a Series.\n",
    "\n",
    "We can do this using `.loc[index_items]`; where `index_items` is\n",
    "an item from the index, or a list of items in the index.\n",
    "\n",
    "We will see this more in-depth in a coming lecture, but for now, we\n",
    "demonstrate how to select one or multiple elements of the Series."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5.6"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp.loc[1995]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1995    5.6\n",
       "2005    5.3\n",
       "2015    5.7\n",
       "Name: Unemployment, dtype: float64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp.loc[[1995, 2005, 2015]]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "<a id='exercise-0'></a>\n",
    "> See exercise 1 in the [*exercise list*](#exerciselist-0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## DataFrame\n",
    "\n",
    "A DataFrame is how pandas stores one or more columns of data.\n",
    "\n",
    "We can think a DataFrames a multiple Series stacked side by side as\n",
    "columns.\n",
    "\n",
    "This is similar to a sheet in an Excel workbook or a table in a SQL\n",
    "database.\n",
    "\n",
    "In addition to row labels (an index), DataFrames also have column labels.\n",
    "\n",
    "We refer to these column labels as the columns or column names.\n",
    "\n",
    "<img src=\"https://datascience.quantecon.org/assets/_static/intro_files/PandasDataFrame.png\" alt=\"PandasDataFrame.png\" style=\"\">\n",
    "\n",
    "  \n",
    "Below, we create a DataFrame that contains the unemployment rate every\n",
    "other year by region of the US starting in 1995."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NorthEast</th>\n",
       "      <th>MidWest</th>\n",
       "      <th>South</th>\n",
       "      <th>West</th>\n",
       "      <th>National</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1995</th>\n",
       "      <td>5.9</td>\n",
       "      <td>4.5</td>\n",
       "      <td>5.3</td>\n",
       "      <td>6.6</td>\n",
       "      <td>5.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1997</th>\n",
       "      <td>5.6</td>\n",
       "      <td>4.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>6.0</td>\n",
       "      <td>5.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1999</th>\n",
       "      <td>4.4</td>\n",
       "      <td>3.6</td>\n",
       "      <td>4.2</td>\n",
       "      <td>5.2</td>\n",
       "      <td>4.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001</th>\n",
       "      <td>3.8</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.6</td>\n",
       "      <td>4.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003</th>\n",
       "      <td>5.8</td>\n",
       "      <td>5.7</td>\n",
       "      <td>5.7</td>\n",
       "      <td>6.5</td>\n",
       "      <td>5.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005</th>\n",
       "      <td>4.9</td>\n",
       "      <td>5.7</td>\n",
       "      <td>5.2</td>\n",
       "      <td>5.5</td>\n",
       "      <td>5.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007</th>\n",
       "      <td>4.3</td>\n",
       "      <td>4.9</td>\n",
       "      <td>4.3</td>\n",
       "      <td>4.5</td>\n",
       "      <td>4.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2009</th>\n",
       "      <td>7.1</td>\n",
       "      <td>8.1</td>\n",
       "      <td>7.6</td>\n",
       "      <td>8.6</td>\n",
       "      <td>7.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011</th>\n",
       "      <td>8.3</td>\n",
       "      <td>8.7</td>\n",
       "      <td>9.1</td>\n",
       "      <td>10.7</td>\n",
       "      <td>9.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013</th>\n",
       "      <td>7.9</td>\n",
       "      <td>7.4</td>\n",
       "      <td>7.4</td>\n",
       "      <td>8.5</td>\n",
       "      <td>8.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015</th>\n",
       "      <td>5.7</td>\n",
       "      <td>5.1</td>\n",
       "      <td>5.5</td>\n",
       "      <td>6.1</td>\n",
       "      <td>5.7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      NorthEast  MidWest  South  West  National\n",
       "1995        5.9      4.5    5.3   6.6       5.6\n",
       "1997        5.6      4.3    5.2   6.0       5.3\n",
       "1999        4.4      3.6    4.2   5.2       4.3\n",
       "2001        3.8      4.0    4.0   4.6       4.2\n",
       "2003        5.8      5.7    5.7   6.5       5.8\n",
       "2005        4.9      5.7    5.2   5.5       5.3\n",
       "2007        4.3      4.9    4.3   4.5       4.6\n",
       "2009        7.1      8.1    7.6   8.6       7.8\n",
       "2011        8.3      8.7    9.1  10.7       9.1\n",
       "2013        7.9      7.4    7.4   8.5       8.0\n",
       "2015        5.7      5.1    5.5   6.1       5.7"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = {\n",
    "    \"NorthEast\": [5.9,  5.6,  4.4,  3.8,  5.8,  4.9,  4.3,  7.1,  8.3,  7.9,  5.7],\n",
    "    \"MidWest\": [4.5,  4.3,  3.6,  4. ,  5.7,  5.7,  4.9,  8.1,  8.7,  7.4,  5.1],\n",
    "    \"South\": [5.3,  5.2,  4.2,  4. ,  5.7,  5.2,  4.3,  7.6,  9.1,  7.4,  5.5],\n",
    "    \"West\": [6.6, 6., 5.2, 4.6, 6.5, 5.5, 4.5, 8.6, 10.7, 8.5, 6.1],\n",
    "    \"National\": [5.6, 5.3, 4.3, 4.2, 5.8, 5.3, 4.6, 7.8, 9.1, 8., 5.7]\n",
    "}\n",
    "\n",
    "unemp_region = pd.DataFrame(data, index=years)\n",
    "unemp_region"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can retrieve the index and the DataFrame values as we\n",
    "did with a Series."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Int64Index([1995, 1997, 1999, 2001, 2003, 2005, 2007, 2009, 2011, 2013, 2015], dtype='int64')"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp_region.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 5.9,  4.5,  5.3,  6.6,  5.6],\n",
       "       [ 5.6,  4.3,  5.2,  6. ,  5.3],\n",
       "       [ 4.4,  3.6,  4.2,  5.2,  4.3],\n",
       "       [ 3.8,  4. ,  4. ,  4.6,  4.2],\n",
       "       [ 5.8,  5.7,  5.7,  6.5,  5.8],\n",
       "       [ 4.9,  5.7,  5.2,  5.5,  5.3],\n",
       "       [ 4.3,  4.9,  4.3,  4.5,  4.6],\n",
       "       [ 7.1,  8.1,  7.6,  8.6,  7.8],\n",
       "       [ 8.3,  8.7,  9.1, 10.7,  9.1],\n",
       "       [ 7.9,  7.4,  7.4,  8.5,  8. ],\n",
       "       [ 5.7,  5.1,  5.5,  6.1,  5.7]])"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp_region.values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### What Can We Do with a DataFrame?\n",
    "\n",
    "Pretty much everything we can do with a Series."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### `.head` and `.tail`\n",
    "\n",
    "As with Series, we can use `.head` and `.tail` to show only the\n",
    "first or last `n` rows."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NorthEast</th>\n",
       "      <th>MidWest</th>\n",
       "      <th>South</th>\n",
       "      <th>West</th>\n",
       "      <th>National</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1995</th>\n",
       "      <td>5.9</td>\n",
       "      <td>4.5</td>\n",
       "      <td>5.3</td>\n",
       "      <td>6.6</td>\n",
       "      <td>5.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1997</th>\n",
       "      <td>5.6</td>\n",
       "      <td>4.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>6.0</td>\n",
       "      <td>5.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1999</th>\n",
       "      <td>4.4</td>\n",
       "      <td>3.6</td>\n",
       "      <td>4.2</td>\n",
       "      <td>5.2</td>\n",
       "      <td>4.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001</th>\n",
       "      <td>3.8</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.6</td>\n",
       "      <td>4.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003</th>\n",
       "      <td>5.8</td>\n",
       "      <td>5.7</td>\n",
       "      <td>5.7</td>\n",
       "      <td>6.5</td>\n",
       "      <td>5.8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      NorthEast  MidWest  South  West  National\n",
       "1995        5.9      4.5    5.3   6.6       5.6\n",
       "1997        5.6      4.3    5.2   6.0       5.3\n",
       "1999        4.4      3.6    4.2   5.2       4.3\n",
       "2001        3.8      4.0    4.0   4.6       4.2\n",
       "2003        5.8      5.7    5.7   6.5       5.8"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp_region.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NorthEast</th>\n",
       "      <th>MidWest</th>\n",
       "      <th>South</th>\n",
       "      <th>West</th>\n",
       "      <th>National</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2011</th>\n",
       "      <td>8.3</td>\n",
       "      <td>8.7</td>\n",
       "      <td>9.1</td>\n",
       "      <td>10.7</td>\n",
       "      <td>9.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013</th>\n",
       "      <td>7.9</td>\n",
       "      <td>7.4</td>\n",
       "      <td>7.4</td>\n",
       "      <td>8.5</td>\n",
       "      <td>8.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015</th>\n",
       "      <td>5.7</td>\n",
       "      <td>5.1</td>\n",
       "      <td>5.5</td>\n",
       "      <td>6.1</td>\n",
       "      <td>5.7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      NorthEast  MidWest  South  West  National\n",
       "2011        8.3      8.7    9.1  10.7       9.1\n",
       "2013        7.9      7.4    7.4   8.5       8.0\n",
       "2015        5.7      5.1    5.5   6.1       5.7"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp_region.tail(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Plotting\n",
    "\n",
    "We can generate plots with the `.plot` method.\n",
    "\n",
    "Notice we now have a separate line for each column of data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7ff4b78dc278>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "unemp_region.plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Indexing\n",
    "\n",
    "We can also do indexing using `.loc`.\n",
    "\n",
    "This is slightly more advanced than before because we can choose\n",
    "subsets of both row and columns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5.9"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp_region.loc[1995, \"NorthEast\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1995    5.3\n",
       "2005    5.2\n",
       "Name: South, dtype: float64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp_region.loc[[1995, 2005], \"South\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NorthEast    5.9\n",
       "National     5.6\n",
       "Name: 1995, dtype: float64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp_region.loc[1995, [\"NorthEast\", \"National\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1995    5.9\n",
       "1997    5.6\n",
       "1999    4.4\n",
       "2001    3.8\n",
       "2003    5.8\n",
       "2005    4.9\n",
       "2007    4.3\n",
       "2009    7.1\n",
       "2011    8.3\n",
       "2013    7.9\n",
       "2015    5.7\n",
       "Name: NorthEast, dtype: float64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp_region.loc[:, \"NorthEast\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1995    4.5\n",
       "1997    4.3\n",
       "1999    3.6\n",
       "2001    4.0\n",
       "2003    5.7\n",
       "2005    5.7\n",
       "2007    4.9\n",
       "2009    8.1\n",
       "2011    8.7\n",
       "2013    7.4\n",
       "2015    5.1\n",
       "Name: MidWest, dtype: float64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# `[string]` with no `.loc` extracts a whole column\n",
    "unemp_region[\"MidWest\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Computations with Columns\n",
    "\n",
    "pandas can do various computations and mathematical operations on\n",
    "columns.\n",
    "\n",
    "Let’s take a look at a few of them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1995    0.066\n",
       "1997    0.060\n",
       "1999    0.052\n",
       "2001    0.046\n",
       "2003    0.065\n",
       "2005    0.055\n",
       "2007    0.045\n",
       "2009    0.086\n",
       "2011    0.107\n",
       "2013    0.085\n",
       "2015    0.061\n",
       "Name: West, dtype: float64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Divide by 100 to move from percent units to a rate\n",
    "unemp_region[\"West\"] / 100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10.7"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Find maximum\n",
    "unemp_region[\"West\"].max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1995    2.1\n",
       "1997    1.7\n",
       "1999    1.6\n",
       "2001    0.6\n",
       "2003    0.8\n",
       "2005   -0.2\n",
       "2007   -0.4\n",
       "2009    0.5\n",
       "2011    2.0\n",
       "2013    1.1\n",
       "2015    1.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Find the difference between two columns\n",
    "# Notice that pandas applies `-` to _all rows_ at once\n",
    "# We'll see more of this throughout these materials\n",
    "unemp_region[\"West\"] - unemp_region[\"MidWest\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9006381255384481"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Find correlation between two columns\n",
    "unemp_region.West.corr(unemp_region[\"MidWest\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NorthEast</th>\n",
       "      <th>MidWest</th>\n",
       "      <th>South</th>\n",
       "      <th>West</th>\n",
       "      <th>National</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>NorthEast</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.875654</td>\n",
       "      <td>0.964415</td>\n",
       "      <td>0.967875</td>\n",
       "      <td>0.976016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MidWest</th>\n",
       "      <td>0.875654</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.951379</td>\n",
       "      <td>0.900638</td>\n",
       "      <td>0.952389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>South</th>\n",
       "      <td>0.964415</td>\n",
       "      <td>0.951379</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.987259</td>\n",
       "      <td>0.995030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>West</th>\n",
       "      <td>0.967875</td>\n",
       "      <td>0.900638</td>\n",
       "      <td>0.987259</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.981308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>National</th>\n",
       "      <td>0.976016</td>\n",
       "      <td>0.952389</td>\n",
       "      <td>0.995030</td>\n",
       "      <td>0.981308</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           NorthEast   MidWest     South      West  National\n",
       "NorthEast   1.000000  0.875654  0.964415  0.967875  0.976016\n",
       "MidWest     0.875654  1.000000  0.951379  0.900638  0.952389\n",
       "South       0.964415  0.951379  1.000000  0.987259  0.995030\n",
       "West        0.967875  0.900638  0.987259  1.000000  0.981308\n",
       "National    0.976016  0.952389  0.995030  0.981308  1.000000"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# find correlation between all column pairs\n",
    "unemp_region.corr()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "<a id='exercise-1'></a>\n",
    "> See exercise 2 in the [*exercise list*](#exerciselist-0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Types\n",
    "\n",
    "We asked you to run the commands `unemp.dtype` and\n",
    "`unemp_region.dtypes` and think about the outputs.\n",
    "\n",
    "You might have guessed that they return the type of the values inside\n",
    "each column.\n",
    "\n",
    "Occasionally, you might need to investigate what types you have in your\n",
    "DataFrame when an operation isn’t behaving as expected."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtype('float64')"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp.dtype"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NorthEast    float64\n",
       "MidWest      float64\n",
       "South        float64\n",
       "West         float64\n",
       "National     float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp_region.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "DataFrames will only distinguish between a few types.\n",
    "\n",
    "- Booleans (`bool`)  \n",
    "- Floating point numbers (`float64`)  \n",
    "- Integers (`int64`)  \n",
    "- Dates (`datetime`) — we will learn this soon  \n",
    "- Categorical data (`categorical`)  \n",
    "- Everything else, including strings (`object`)  \n",
    "\n",
    "\n",
    "In the future, we will often refer to the type of data stored in a\n",
    "column as its `dtype`.\n",
    "\n",
    "Let’s look at an example for when having an incorrect `dtype` can\n",
    "cause problems.\n",
    "\n",
    "Suppose that when we imported the data the `South` column was\n",
    "interpreted as a string."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NorthEast    float64\n",
       "MidWest      float64\n",
       "South         object\n",
       "West         float64\n",
       "National     float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "str_unemp = unemp_region.copy()\n",
    "str_unemp[\"South\"] = str_unemp[\"South\"].astype(str)\n",
    "str_unemp.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Everything *looks* ok…"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NorthEast</th>\n",
       "      <th>MidWest</th>\n",
       "      <th>South</th>\n",
       "      <th>West</th>\n",
       "      <th>National</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1995</th>\n",
       "      <td>5.9</td>\n",
       "      <td>4.5</td>\n",
       "      <td>5.3</td>\n",
       "      <td>6.6</td>\n",
       "      <td>5.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1997</th>\n",
       "      <td>5.6</td>\n",
       "      <td>4.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>6.0</td>\n",
       "      <td>5.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1999</th>\n",
       "      <td>4.4</td>\n",
       "      <td>3.6</td>\n",
       "      <td>4.2</td>\n",
       "      <td>5.2</td>\n",
       "      <td>4.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001</th>\n",
       "      <td>3.8</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.6</td>\n",
       "      <td>4.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003</th>\n",
       "      <td>5.8</td>\n",
       "      <td>5.7</td>\n",
       "      <td>5.7</td>\n",
       "      <td>6.5</td>\n",
       "      <td>5.8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      NorthEast  MidWest South  West  National\n",
       "1995        5.9      4.5   5.3   6.6       5.6\n",
       "1997        5.6      4.3   5.2   6.0       5.3\n",
       "1999        4.4      3.6   4.2   5.2       4.3\n",
       "2001        3.8      4.0   4.0   4.6       4.2\n",
       "2003        5.8      5.7   5.7   6.5       5.8"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "str_unemp.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "But if we try to do something like compute the sum of all the columns,\n",
    "we get unexpected results…"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NorthEast                                 63.7\n",
       "MidWest                                     62\n",
       "South        5.35.24.24.05.75.24.37.69.17.45.5\n",
       "West                                      72.8\n",
       "National                                  65.7\n",
       "dtype: object"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "str_unemp.sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This happened because `.sum` effectively calls `+` on all rows in\n",
    "each column.\n",
    "\n",
    "Recall that when we apply `+` to two strings, the result is the two\n",
    "strings concatenated.\n",
    "\n",
    "So, in this case, we saw that the entries in all rows of the South\n",
    "column were stitched together into one long string."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Changing DataFrames\n",
    "\n",
    "We can change the data inside of a DataFrame in various ways:\n",
    "\n",
    "- Adding new columns  \n",
    "- Changing index labels or column names  \n",
    "- Altering existing data (e.g. doing some arithmetic or making a column\n",
    "  of strings lowercase)  \n",
    "\n",
    "\n",
    "Some of these “mutations” will be topics of future lectures, so we will\n",
    "only briefly discuss a few of the things we can do below."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Creating New Columns\n",
    "\n",
    "We can create new data by assigning values to a column similar to how\n",
    "we assign values to a variable.\n",
    "\n",
    "In pandas, we create a new column of a DataFrame by writing:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hide-output": false
   },
   "source": [
    "```python\n",
    "df[\"New Column Name\"] = new_values\n",
    "```\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Below, we create an unweighted mean of the unemployment rate across the\n",
    "four regions of the US — notice that this differs from the national\n",
    "unemployment rate."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "hide-output": false
   },
   "outputs": [],
   "source": [
    "unemp_region[\"UnweightedMean\"] = (unemp_region[\"NorthEast\"] +\n",
    "                                  unemp_region[\"MidWest\"] +\n",
    "                                  unemp_region[\"South\"] +\n",
    "                                  unemp_region[\"West\"])/4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NorthEast</th>\n",
       "      <th>MidWest</th>\n",
       "      <th>South</th>\n",
       "      <th>West</th>\n",
       "      <th>National</th>\n",
       "      <th>UnweightedMean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1995</th>\n",
       "      <td>5.9</td>\n",
       "      <td>4.5</td>\n",
       "      <td>5.3</td>\n",
       "      <td>6.6</td>\n",
       "      <td>5.6</td>\n",
       "      <td>5.575</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1997</th>\n",
       "      <td>5.6</td>\n",
       "      <td>4.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>6.0</td>\n",
       "      <td>5.3</td>\n",
       "      <td>5.275</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1999</th>\n",
       "      <td>4.4</td>\n",
       "      <td>3.6</td>\n",
       "      <td>4.2</td>\n",
       "      <td>5.2</td>\n",
       "      <td>4.3</td>\n",
       "      <td>4.350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001</th>\n",
       "      <td>3.8</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.6</td>\n",
       "      <td>4.2</td>\n",
       "      <td>4.100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003</th>\n",
       "      <td>5.8</td>\n",
       "      <td>5.7</td>\n",
       "      <td>5.7</td>\n",
       "      <td>6.5</td>\n",
       "      <td>5.8</td>\n",
       "      <td>5.925</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      NorthEast  MidWest  South  West  National  UnweightedMean\n",
       "1995        5.9      4.5    5.3   6.6       5.6           5.575\n",
       "1997        5.6      4.3    5.2   6.0       5.3           5.275\n",
       "1999        4.4      3.6    4.2   5.2       4.3           4.350\n",
       "2001        3.8      4.0    4.0   4.6       4.2           4.100\n",
       "2003        5.8      5.7    5.7   6.5       5.8           5.925"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp_region.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Changing Values\n",
    "\n",
    "Changing the values inside of a DataFrame should be done sparingly.\n",
    "\n",
    "However, it can be done by assigning a value to a location in the\n",
    "DataFrame.\n",
    "\n",
    "`df.loc[index, column] = value`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "hide-output": false
   },
   "outputs": [],
   "source": [
    "unemp_region.loc[1995, \"UnweightedMean\"] = 0.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NorthEast</th>\n",
       "      <th>MidWest</th>\n",
       "      <th>South</th>\n",
       "      <th>West</th>\n",
       "      <th>National</th>\n",
       "      <th>UnweightedMean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1995</th>\n",
       "      <td>5.9</td>\n",
       "      <td>4.5</td>\n",
       "      <td>5.3</td>\n",
       "      <td>6.6</td>\n",
       "      <td>5.6</td>\n",
       "      <td>0.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1997</th>\n",
       "      <td>5.6</td>\n",
       "      <td>4.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>6.0</td>\n",
       "      <td>5.3</td>\n",
       "      <td>5.275</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1999</th>\n",
       "      <td>4.4</td>\n",
       "      <td>3.6</td>\n",
       "      <td>4.2</td>\n",
       "      <td>5.2</td>\n",
       "      <td>4.3</td>\n",
       "      <td>4.350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001</th>\n",
       "      <td>3.8</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.6</td>\n",
       "      <td>4.2</td>\n",
       "      <td>4.100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003</th>\n",
       "      <td>5.8</td>\n",
       "      <td>5.7</td>\n",
       "      <td>5.7</td>\n",
       "      <td>6.5</td>\n",
       "      <td>5.8</td>\n",
       "      <td>5.925</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      NorthEast  MidWest  South  West  National  UnweightedMean\n",
       "1995        5.9      4.5    5.3   6.6       5.6           0.000\n",
       "1997        5.6      4.3    5.2   6.0       5.3           5.275\n",
       "1999        4.4      3.6    4.2   5.2       4.3           4.350\n",
       "2001        3.8      4.0    4.0   4.6       4.2           4.100\n",
       "2003        5.8      5.7    5.7   6.5       5.8           5.925"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp_region.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Renaming Columns\n",
    "\n",
    "We can also rename the columns of a DataFrame, which is helpful because the names that sometimes come with datasets are\n",
    "unbearable…\n",
    "\n",
    "For example, the original name for the North East unemployment rate\n",
    "given by the Bureau of Labor Statistics was `LASRD910000000000003`…\n",
    "\n",
    "They have their reasons for using these names, but it can make our job\n",
    "difficult since we often need to type it repeatedly.\n",
    "\n",
    "We can rename columns by passing a dictionary to the `rename` method.\n",
    "\n",
    "This dictionary contains the old names as the keys and new names as the\n",
    "values.\n",
    "\n",
    "See the example below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NE</th>\n",
       "      <th>MW</th>\n",
       "      <th>S</th>\n",
       "      <th>W</th>\n",
       "      <th>National</th>\n",
       "      <th>UnweightedMean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1995</th>\n",
       "      <td>5.9</td>\n",
       "      <td>4.5</td>\n",
       "      <td>5.3</td>\n",
       "      <td>6.6</td>\n",
       "      <td>5.6</td>\n",
       "      <td>0.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1997</th>\n",
       "      <td>5.6</td>\n",
       "      <td>4.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>6.0</td>\n",
       "      <td>5.3</td>\n",
       "      <td>5.275</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1999</th>\n",
       "      <td>4.4</td>\n",
       "      <td>3.6</td>\n",
       "      <td>4.2</td>\n",
       "      <td>5.2</td>\n",
       "      <td>4.3</td>\n",
       "      <td>4.350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001</th>\n",
       "      <td>3.8</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.6</td>\n",
       "      <td>4.2</td>\n",
       "      <td>4.100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003</th>\n",
       "      <td>5.8</td>\n",
       "      <td>5.7</td>\n",
       "      <td>5.7</td>\n",
       "      <td>6.5</td>\n",
       "      <td>5.8</td>\n",
       "      <td>5.925</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005</th>\n",
       "      <td>4.9</td>\n",
       "      <td>5.7</td>\n",
       "      <td>5.2</td>\n",
       "      <td>5.5</td>\n",
       "      <td>5.3</td>\n",
       "      <td>5.325</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007</th>\n",
       "      <td>4.3</td>\n",
       "      <td>4.9</td>\n",
       "      <td>4.3</td>\n",
       "      <td>4.5</td>\n",
       "      <td>4.6</td>\n",
       "      <td>4.500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2009</th>\n",
       "      <td>7.1</td>\n",
       "      <td>8.1</td>\n",
       "      <td>7.6</td>\n",
       "      <td>8.6</td>\n",
       "      <td>7.8</td>\n",
       "      <td>7.850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011</th>\n",
       "      <td>8.3</td>\n",
       "      <td>8.7</td>\n",
       "      <td>9.1</td>\n",
       "      <td>10.7</td>\n",
       "      <td>9.1</td>\n",
       "      <td>9.200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013</th>\n",
       "      <td>7.9</td>\n",
       "      <td>7.4</td>\n",
       "      <td>7.4</td>\n",
       "      <td>8.5</td>\n",
       "      <td>8.0</td>\n",
       "      <td>7.800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015</th>\n",
       "      <td>5.7</td>\n",
       "      <td>5.1</td>\n",
       "      <td>5.5</td>\n",
       "      <td>6.1</td>\n",
       "      <td>5.7</td>\n",
       "      <td>5.600</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       NE   MW    S     W  National  UnweightedMean\n",
       "1995  5.9  4.5  5.3   6.6       5.6           0.000\n",
       "1997  5.6  4.3  5.2   6.0       5.3           5.275\n",
       "1999  4.4  3.6  4.2   5.2       4.3           4.350\n",
       "2001  3.8  4.0  4.0   4.6       4.2           4.100\n",
       "2003  5.8  5.7  5.7   6.5       5.8           5.925\n",
       "2005  4.9  5.7  5.2   5.5       5.3           5.325\n",
       "2007  4.3  4.9  4.3   4.5       4.6           4.500\n",
       "2009  7.1  8.1  7.6   8.6       7.8           7.850\n",
       "2011  8.3  8.7  9.1  10.7       9.1           9.200\n",
       "2013  7.9  7.4  7.4   8.5       8.0           7.800\n",
       "2015  5.7  5.1  5.5   6.1       5.7           5.600"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "names = {\"NorthEast\": \"NE\",\n",
    "         \"MidWest\": \"MW\",\n",
    "         \"South\": \"S\",\n",
    "         \"West\": \"W\"}\n",
    "unemp_region.rename(columns=names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NorthEast</th>\n",
       "      <th>MidWest</th>\n",
       "      <th>South</th>\n",
       "      <th>West</th>\n",
       "      <th>National</th>\n",
       "      <th>UnweightedMean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1995</th>\n",
       "      <td>5.9</td>\n",
       "      <td>4.5</td>\n",
       "      <td>5.3</td>\n",
       "      <td>6.6</td>\n",
       "      <td>5.6</td>\n",
       "      <td>0.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1997</th>\n",
       "      <td>5.6</td>\n",
       "      <td>4.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>6.0</td>\n",
       "      <td>5.3</td>\n",
       "      <td>5.275</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1999</th>\n",
       "      <td>4.4</td>\n",
       "      <td>3.6</td>\n",
       "      <td>4.2</td>\n",
       "      <td>5.2</td>\n",
       "      <td>4.3</td>\n",
       "      <td>4.350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001</th>\n",
       "      <td>3.8</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.6</td>\n",
       "      <td>4.2</td>\n",
       "      <td>4.100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003</th>\n",
       "      <td>5.8</td>\n",
       "      <td>5.7</td>\n",
       "      <td>5.7</td>\n",
       "      <td>6.5</td>\n",
       "      <td>5.8</td>\n",
       "      <td>5.925</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      NorthEast  MidWest  South  West  National  UnweightedMean\n",
       "1995        5.9      4.5    5.3   6.6       5.6           0.000\n",
       "1997        5.6      4.3    5.2   6.0       5.3           5.275\n",
       "1999        4.4      3.6    4.2   5.2       4.3           4.350\n",
       "2001        3.8      4.0    4.0   4.6       4.2           4.100\n",
       "2003        5.8      5.7    5.7   6.5       5.8           5.925"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unemp_region.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We renamed our columns… Why does the DataFrame still show the old\n",
    "column names?\n",
    "\n",
    "Many pandas operations create a copy of your data by\n",
    "default to protect your data and prevent you from overwriting\n",
    "information you meant to keep.\n",
    "\n",
    "We can make these operations permanent by either:\n",
    "\n",
    "1. Assigning the output back to the variable name\n",
    "  `df = df.rename(columns=rename_dict)`  \n",
    "1. Looking into whether the method has an `inplace` option. For\n",
    "  example, `df.rename(columns=rename_dict, inplace=True)`  \n",
    "\n",
    "\n",
    "Setting `inplace=True` will sometimes make your code faster\n",
    "(e.g. if you have a very large DataFrame and you don’t want to copy all\n",
    "the data), but that doesn’t always happen.\n",
    "\n",
    "We recommend using the first option until you get comfortable with\n",
    "pandas because operations that don’t alter your data are (usually)\n",
    "safer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "hide-output": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NE</th>\n",
       "      <th>MW</th>\n",
       "      <th>S</th>\n",
       "      <th>W</th>\n",
       "      <th>National</th>\n",
       "      <th>UnweightedMean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1995</th>\n",
       "      <td>5.9</td>\n",
       "      <td>4.5</td>\n",
       "      <td>5.3</td>\n",
       "      <td>6.6</td>\n",
       "      <td>5.6</td>\n",
       "      <td>0.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1997</th>\n",
       "      <td>5.6</td>\n",
       "      <td>4.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>6.0</td>\n",
       "      <td>5.3</td>\n",
       "      <td>5.275</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1999</th>\n",
       "      <td>4.4</td>\n",
       "      <td>3.6</td>\n",
       "      <td>4.2</td>\n",
       "      <td>5.2</td>\n",
       "      <td>4.3</td>\n",
       "      <td>4.350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001</th>\n",
       "      <td>3.8</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.6</td>\n",
       "      <td>4.2</td>\n",
       "      <td>4.100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003</th>\n",
       "      <td>5.8</td>\n",
       "      <td>5.7</td>\n",
       "      <td>5.7</td>\n",
       "      <td>6.5</td>\n",
       "      <td>5.8</td>\n",
       "      <td>5.925</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       NE   MW    S    W  National  UnweightedMean\n",
       "1995  5.9  4.5  5.3  6.6       5.6           0.000\n",
       "1997  5.6  4.3  5.2  6.0       5.3           5.275\n",
       "1999  4.4  3.6  4.2  5.2       4.3           4.350\n",
       "2001  3.8  4.0  4.0  4.6       4.2           4.100\n",
       "2003  5.8  5.7  5.7  6.5       5.8           5.925"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "names = {\"NorthEast\": \"NE\",\n",
    "         \"MidWest\": \"MW\",\n",
    "         \"South\": \"S\",\n",
    "         \"West\": \"W\"}\n",
    "\n",
    "unemp_shortname = unemp_region.rename(columns=names)\n",
    "unemp_shortname.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exercises\n",
    "\n",
    "\n",
    "<a id='exerciselist-0'></a>\n",
    "**Exercise 1**\n",
    "\n",
    "For each of the following exercises, we recommend reading the documentation\n",
    "for help.\n",
    "\n",
    "- Display only the first 2 elements of the Series using the `.head` method.  \n",
    "- Using the `plot` method, make a bar plot.  \n",
    "- Use `.loc` to select the lowest/highest unemployment rate shown in the Series.  \n",
    "- Run the code `unemp.dtype` below. What does it give you? Where do you think it comes from?  \n",
    "\n",
    "\n",
    "([*back to text*](#exercise-0))\n",
    "\n",
    "**Exercise 2**\n",
    "\n",
    "For each of the following, we recommend reading the documentation for help.\n",
    "\n",
    "- Use introspection (or google-fu) to find a way to obtain a list with\n",
    "  all of the column names in `unemp_region`.  \n",
    "- Using the `plot` method, make a bar plot. What does it look like\n",
    "  now?  \n",
    "- Use `.loc` to select the the unemployment data for the\n",
    "  `NorthEast` and `West` for the years 1995, 2005, 2011, and 2015.  \n",
    "- Run the code `unemp_region.dtypes` below. What does it give you?\n",
    "  How does this compare with `unemp.dtype`?  \n",
    "\n",
    "\n",
    "([*back to text*](#exercise-1))"
   ]
  }
 ],
 "metadata": {
  "date": 1595352472.3606322,
  "download_nb": false,
  "filename": "intro.rst",
  "filename_with_path": "pandas/intro",
  "kernelspec": {
   "display_name": "Python",
   "language": "python3",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  },
  "title": "Introduction"
 },
 "nbformat": 4,
 "nbformat_minor": 2
}