Print the first five rows of the gapminder DataFrame

<tr style="text-align: right;">
  <th></th>
  <th>country</th>
  <th>continent</th>
  <th>year</th>
  <th>lifeExp</th>
  <th>pop</th>
  <th>gdpPercap</th>
</tr>

<tr>
  <th>0</th>
  <td>Afghanistan</td>
  <td>Asia</td>
  <td>1952</td>
  <td>28.801</td>
  <td>8425333</td>
  <td>779.445314</td>
</tr>
<tr>
  <th>1</th>
  <td>Afghanistan</td>
  <td>Asia</td>
  <td>1957</td>
  <td>30.332</td>
  <td>9240934</td>
  <td>820.853030</td>
</tr>
<tr>
  <th>2</th>
  <td>Afghanistan</td>
  <td>Asia</td>
  <td>1962</td>
  <td>31.997</td>
  <td>10267083</td>
  <td>853.100710</td>
</tr>
<tr>
  <th>3</th>
  <td>Afghanistan</td>
  <td>Asia</td>
  <td>1967</td>
  <td>34.020</td>
  <td>11537966</td>
  <td>836.197138</td>
</tr>
<tr>
  <th>4</th>
  <td>Afghanistan</td>
  <td>Asia</td>
  <td>1972</td>
  <td>36.088</td>
  <td>13079460</td>
  <td>739.981106</td>
</tr>

Generate the same graph as above, but this time log-transform the population variable

<matplotlib.axes._subplots.AxesSubplot at 0x108450518>

png

(To Assess the Relationship Between Year and Life Expectancy for a Given Country)

Here the goal is to write a basic function, “life_expectancy”, that incorporates your work above.

By default, the function should return a scatterplot of life-expectancy versus years for a given country. [Hint: Subset the data for a specific country, similar to a problem above]

Once you subset the data, the function should do one of two things: * (1) return a graph [or] * (2) return a graph and model results.

Thus, your function should have arguments and output as follows:

*   Arguments:
        Country (required):   The name of a specific country, such as "Australia"
        Model   (optional):   Build and Return a Model Results, #Hint, set the default to be False

*   Output: 
        (1) - Default: A scatterplot of the relationship with best fit line
        (2) - Model:   The above graph AND the model results

Of course, we have not yet covered modeling in Python.

To run a linear model, we can use the library statsmodels, to predict life expectancy by year.

Example for this problem:

import statsmodels.formula.api as sm #Import Package
model = sm.ols(formula = 'lifeExp ~ year', data = gapminder).fit() #Fit OLS Model
results = model.summary() #Get Results
print(results) # Print

#Hint: Use this Code in Your Function. 
#You will need to replace data = gapminder, with the data subset for a specific country.

                            OLS Regression Results                            
==============================================================================
Dep. Variable:                lifeExp   R-squared:                       0.190
Model:                            OLS   Adj. R-squared:                  0.189
Method:                 Least Squares   F-statistic:                     398.6
Date:                Tue, 25 Oct 2016   Prob (F-statistic):           7.55e-80
Time:                        16:40:53   Log-Likelihood:                -6597.9
No. Observations:                1704   AIC:                         1.320e+04
Df Residuals:                    1702   BIC:                         1.321e+04
Df Model:                           1                                         
Covariance Type:            nonrobust                                         
==============================================================================
                 coef    std err          t      P>|t|      [95.0% Conf. Int.]
------------------------------------------------------------------------------
Intercept   -585.6522     32.314    -18.124      0.000      -649.031  -522.273
year           0.3259      0.016     19.965      0.000         0.294     0.358
==============================================================================
Omnibus:                      386.124   Durbin-Watson:                   1.962
Prob(Omnibus):                  0.000   Jarque-Bera (JB):               90.750
Skew:                          -0.268   Prob(JB):                     1.97e-20
Kurtosis:                       2.004   Cond. No.                     2.27e+05
==============================================================================

Warnings:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
[2] The condition number is large, 2.27e+05. This might indicate that there are
strong multicollinearity or other numerical problems.

# write your function here

Assess the relationship between year and life expectancy

As you know already, the general trend is that over time life expectancy increases, but the trend is different for each country. Some experience a greater increase than others, whereas some countries experience declines in life expectancy. You can use whatever method you wish to assess and explain this relationship using Python.

• You could draw a graph
• You could draw a graph which visualizes the differing relationships between countries
• You could estimate a correlation coefficient
• You could estimate a statistical model - note that the notebook as written uses Python 2. For the most part it works with Python 3 though.
• You could estimate a statistical model for each country

Use whichever method you think you can master before the assignment is due. Some of you may just stick to basic graphs and tables, while others might build a statistical model using statsmodel. Obviously the more advanced technique you use, the higher your ceiling will be for your evaluation. But don’t spend 10 hours getting this to work! Go with what you can accomplish in a reasonable amount of time.