Supporting Statistical Analysis for Research

5.4 Factors and Indicators

These exercises use the mtcars.csv data set.

1. Import the mtcars.csv data set.

   from pathlib import Path
   import pandas as pd
   import numpy as np

   mtcars_path = Path('..') / 'datasets' / 'mtcars.csv'
   mtcars_in = pd.read_csv(mtcars_path)
   mtcars_in = mtcars_in.rename(columns={'Unnamed: 0': 'make_model'})
   mtcars =  mtcars_in.copy(deep=True)
   
   print(mtcars.dtypes)

   make_model     object
   mpg           float64
   cyl             int64
   disp          float64
   hp              int64
   drat          float64
   wt            float64
   qsec          float64
   vs              int64
   am              int64
   gear            int64
   carb            int64
   dtype: object

2. Factor the cyl, gear and carb variables.

   mtcars = (
       mtcars
           .apply(
               func=lambda x: x.astype('category')
               if x.name in ['cyl', 'gear', 'carb'] else x))
   
   print(mtcars.dtypes)

   make_model      object
   mpg            float64
   cyl           category
   disp           float64
   hp               int64
   drat           float64
   wt             float64
   qsec           float64
   vs               int64
   am               int64
   gear          category
   carb          category
   dtype: object

   or

   mtcars =  mtcars_in.copy(deep=True)
   
   cyl_lev = pd.Series(mtcars['cyl'].unique()).sort_values()
   gear_lev = pd.Series(mtcars['gear'].unique()).sort_values()
   carb_lev = pd.Series(mtcars['carb'].unique()).sort_values()
   mtcars = (
       mtcars
           .assign(
               cyl = lambda df:
                   pd.Categorical(df['cyl'], categories=cyl_lev),
               gear = lambda df:
                   pd.Categorical(df['gear'], categories=gear_lev),
               carb = lambda df:
                   pd.Categorical(df['carb'], categories=carb_lev)))
   
   print(mtcars.dtypes)

   make_model      object
   mpg            float64
   cyl           category
   disp           float64
   hp               int64
   drat           float64
   wt             float64
   qsec           float64
   vs               int64
   am               int64
   gear          category
   carb          category
   dtype: object

3. Create a variable that identifies the observations that are in the top 25 percent of miles per gallon. Display a few of these vehicles.

   Hint, you will need to find a function to identify the percentage points of a variable.

   Note, that the quantile function returns a series.

   mtcars = (
       mtcars
           .assign(
               efficient = lambda df:
                   np.where(
                       df['mpg'] >= df['mpg'].quantile([0.75]).at[0.75],
                       True,
                       False)))
   (mtcars
       .loc[:, ['make_model', 'mpg', 'efficient']]
       .head()
       .pipe(print))

             make_model   mpg  efficient
   0          Mazda RX4  21.0      False
   1      Mazda RX4 Wag  21.0      False
   2         Datsun 710  22.8       True
   3     Hornet 4 Drive  21.4      False
   4  Hornet Sportabout  18.7      False

   or

   mtcars = (
       mtcars
           .assign(
               efficient = lambda df:
                   np.where(
                       df['mpg'] >= df['mpg'].quantile([0.75]).iloc[0],
                       True,
                       False)))
   (mtcars
       .loc[:, ['make_model', 'mpg', 'efficient']]
       .head()
       .pipe(print))

             make_model   mpg  efficient
   0          Mazda RX4  21.0      False
   1      Mazda RX4 Wag  21.0      False
   2         Datsun 710  22.8       True
   3     Hornet 4 Drive  21.4      False
   4  Hornet Sportabout  18.7      False

4. Create a variables that bins the values of hp using the following amounts of hp: 100, 170, 240, and 300.

   mtcars = (
       mtcars
           .assign(
               power = lambda df:
                   pd.cut(df['hp'],
                       bins=[-np.inf, 100, 170, 240, 300, np.inf],
                       labels=['gocart', 'slow', 'typical',
                               'fast', 'beast'])))
   
   (mtcars
       .loc[:, ['make_model', 'mpg', 'efficient', 'power']]
       .head()
       .pipe(print))

             make_model   mpg  efficient    power
   0          Mazda RX4  21.0      False     slow
   1      Mazda RX4 Wag  21.0      False     slow
   2         Datsun 710  22.8       True   gocart
   3     Hornet 4 Drive  21.4      False     slow
   4  Hornet Sportabout  18.7      False  typical