====== General instructions ======

Read following important documentation about:
  * pandas [[http://pandas.pydata.org/pandas-docs/stable/dsintro.html#dataframe|dataframes]]
  * [[http://matplotlib.org/api/pyplot_summary.html|matplotlib.pyplot]]


Get the data from [[http://people.stern.nyu.edu/jsimonof/Casebook/Data/ASCII/README.html|this site]].


Save your scripts in a folder inside the data folder, calling the script folder 'my_scripts' or whaterver. If  'my-scripts' is set as your [[python:generic_features#get_the_current_working_directory_address|current working directory]], then the data files are available under this address '../[data file]', for instantce: '../geyser1.TAB'
\\
====== Eruptions of the "Old Faithful" geyser (p.5) ======

\\

===== Histogram (p.5) =====

<code python>
import pandas as pd
import matplotlib.pyplot as plt
gys1 = pd.DataFrame(pd.read_csv('../geyser1.TAB', '\t'))
g_int = gys1['Interval']
ax = plt.gca()
ax.hist(g_int, bins=20, color='r')
ax.set_xlabel('Intereruption time')
ax.set_ylabel('Frequency')
ax.set_title('Histogram')
plt.show() 
</code>


\\

===== Boxplot (p. 6) =====

<code python>
import matplotlib.pyplot as plt
import pandas as pd
gysr1_boxplot = pd.read_csv('.../geyser1.TAB', '\t')
data_gysr1 = gysr1_boxplot['Interval']
plt.boxplot(data_gysr1)
ax = plt.gca()
ax.set_xlabel('222 cases')
ax.set_ylabel('Interruption time ( minutes')
ax.set_title('Box and Whisker Plot')
plt.show()
</code>


\\

===== ScatterPlot (p. 7) =====

AB: Put face- and edgecolor to change both of them. You can also have two different colors for the in- and outside of each dot.

<code python>
import matplotlib.pyplot as plt
import pandas as pd
geysr1_scatterplot = pd.read_csv('.../geyser1.TAB', '\t')
geysr1_data_Xax = geysr1_scatterplot['Duration']
geysr1_data_Yax = geysr1_scatterplot['Interval']
plt.scatter(geysr1_data_Xax, geysr1_data_Yax, facecolor='y', edgecolor='y')
ax = plt.gca()
ax.set_xlabel('Eruption duration time (minutes)')
ax.set_ylabel('Interuption time (minutes)')
ax.set_title('Scatter Plot of INTERVAL vs DURATION')
plt.show()
</code>


\\


===== Descriptive statistics (p.9) =====

Note: try different examples, e.g. the whole population or only those where 'Duration' <= 3, the whole dataframe

[[http://pandas.pydata.org/pandas-docs/stable/basics.html#descriptive-statistics|doc]] – [[http://www.marsja.se/pandas-python-descriptive-statistics/|example]]

<code python>
import pandas as pd
gysr1 = pd.read_csv('../geyser1.tab', '\t')
gysr1['Duration'][gysr1['Duration'] <= 3].describe()
</code>


\\


===== Boxplot (p.9) =====

Selecting rows in a dataframe: [[http://pandas.pydata.org/pandas-docs/stable/indexing.html#the-where-method-and-masking|doc]] / [[http://stackoverflow.com/questions/17071871/select-rows-from-a-dataframe-based-on-values-in-a-column-in-pandas|example]]

<code python>
import matplotlib.pyplot as plt
import pandas as pd
gysr1 = pd.read_csv('../geyser1.tab', '\t')
gysr1_inf3 = gysr1.loc[gysr1['Duration'] <= 3]
gysr1_sup3 = gysr1.loc[gysr1['Duration'] > 3]
plt.boxplot([gysr1_inf3['Interval'],gysr1_sup3['Interval']], labels= ['inf3','sup3'])
</code>


\\


====== International adoption rates (p.13) ======

===== Boxplot (p.14) =====

<code python>
import matplotlib.pyplot as plt
import pandas as pd
adopt_data = pd.read_csv('D:\Python\Libri\A_Casebook_for_a_First_Course_in_Statistics_and_Data_Analysis_Datasets\Data\Tab\\adopt.TAB', '\t')
adopt1 = adopt_data['Visa91']
plt.boxplot(adopt1)
ax = plt.gca()
ax.set_title('Box and Whisker Plot')
ax.set_xlabel('39 cases')
ax.set_ylabel('Number of visas in 1991')
plt.show()
</code>


\\


===== Histogram (p.14) =====

<code python>
import matplotlib.pyplot as plt
import pandas as pd
adopt_data = pd.read_csv('D:\Python\Libri\A_Casebook_for_a_First_Course_in_Statistics_and_Data_Analysis_Datasets\Data\Tab\\adopt.TAB', '\t')
adopt1 = adopt_data['Visa91']
plt.hist(adopt1)
plt.show()
</code>


\\

=====Histogram with Log(p.18)=====
don't find the way to do it
<code Python>
import pandas as pd
import matplotlib.pyplot as plt
adopt = pd.DataFrame(pd.read_csv('D:\Python\Libri\A_Casebook_for_a_First_Course_in_Statistics_and_Data_Analysis_Datasets\Data\Tab\\adopt.TAB', '\t'))
adopt_loghist = adopt['Visa91']
#adopt_loghist.semilogx() --> was one of the possibilities
ax = plt.gca()
ax.hist(adopt_loghist, bins=10, plt.loglog(0.5,3.5), color='r') #put log=True instead, but you will get the log for the frequencies
plt.gca().set_xscale("log")
ax.set_xlabel('Log (Number of 1991 visas')
ax.set_ylabel('Frequency')
ax.set_title('Histogram')
plt.show() 
</code>


=====Scatterplot (p. 17)=====
<code python>
import matplotlib.pyplot as plt
import pandas as pd
adoption_scatterplot = pd.read_csv('...\adopt.TAB', '\t')
adopt_data_Xax = adoption_scatterplot['Visa88']
adopt_data_Yax = adoption_scatterplot['Visa91']
plt.scatter(adopt_data_Xax, adopt_data_Yax, facecolor='y', edgecolor='y')
ax = plt.gca()
ax.set_xlabel('Number of Visas in 1988')
ax.set_ylim([0,2700])
ax.set_xlim([0,5000])
ax.set_ylabel('Number of Visas in 1991')
ax.set_title('ScatterPlot of Visa91 vs Visa88')
plt.show()
</code>


\\


=====Scatterplot (p.18)=====
<code python>
import matplotlib.pyplot as plt
import pandas as pd
adoption_scatterplot = pd.read_csv('D:\Python\Libri\A_Casebook_for_a_First_Course_in_Statistics_and_Data_Analysis_Datasets\Data\Tab\\adopt.TAB', '\t')
adopt_data_Xax = adoption_scatterplot['Visa91']
adopt_data_Yax = adoption_scatterplot['Visa92']
plt.scatter(adopt_data_Xax, adopt_data_Yax, facecolor='y', edgecolor='y')
ax = plt.gca()
ax.set_xlabel('Number of Visas in 1991')
ax.set_ylim([0,1800])
ax.set_xlim([0,2700])
ax.set_ylabel('Number of Visas in 1992')
ax.set_title('ScatterPlot of Visa92 vs Visa91')
plt.show()
</code>

\\


====== The Performance of stock mutual funds (p. 21) ======






\\

====== Predicting the sales and airplay of popular music (p. 23)======




\\

====== Another look at the "Old faithful" geyser and adoption visas (p.24) ======

Modified the bins of the both histograms:
The Histogram is reliable for the "Old faithful" geyser but not for the Adoption rates. The appearance of the histogram changes quite a lot by changing the bins.

\\

====== Productivity versus quality in the assembly plant (p. 25)======


===== Scatterplot of Productivity vs Quality (p. 26) =====
<code Python>
import pandas as pd
import matplotlib.pyplot as plt
scatter_plot = pd.read_csv('D:\Python\Libri\A_Casebook_for_a_First_Course_in_Statistics_and_Data_Analysis_Datasets\Data\Tab\\prdq.TAB', '\t')
productivity_Y = scatter_plot['Producti']
quality_X = scatter_plot['Quality']
plt.scatter(productivity_Y, quality_X, bins=20, colors='r')
ax = plt.gca()
ax.set_Xlabel('Assembly defects per 100 cars')
ax.set_Ylabel('Hours per vehicle')
ax.set_title('Scatter Plot of Productivity VS Quality')
plt.show()
</code>

\\

=====Scatter Plot of PRODJAPN vs QUALJAPN (p. 27) =====

<code Python>
import pandas as pd
import matplotlib.pyplot as plt
scatter_plot = pd.read_csv('D:\Python\Libri\A_Casebook_for_a_First_Course_in_Statistics_and_Data_Analysis_Datasets\Data\Tab\\prdq.TAB', '\t')
productivity_Y = scatter_plot['ProdJapn']
quality_X = scatter_plot['QualJapn']
plt.scatter(productivity_Y, quality_X, bins=20, colors='r')
ax = plt.gca()
ax.set_Xlabel('Assembly defects per 100 cars (Japanese origin)')
ax.set_Ylabel('Hours per vehicle (Japanese origin')
ax.set_title('Scatter Plot of PRODJAPN VS QUALJAPN')
plt.show()
</code>


=====Scatter Plot of PRODNONJ cs QUALNONJ (p. 27)=====
<code Python>
import pandas as pd
import matplotlib.pyplot as plt
scatter_plot = pd.read_csv('D:\Python\Libri\A_Casebook_for_a_First_Course_in_Statistics_and_Data_Analysis_Datasets\Data\Tab\\prdq.TAB', '\t')
productivity_Y = scatter_plot['ProdNonJ']
quality_X = scatter_plot['QualNonJ']
plt.scatter(productivity_Y, quality_X, bins=20, colors='r')
ax = plt.gca()
ax.set_Xlabel('Assembly defects per 100 cars (non-Japanese origin)')
ax.set_Ylabel('Hours per vehicle (non-Japanese origin')
ax.set_title('Scatter Plot of PRODNONJ VS QUALNONJ')
plt.show()
</code>



===== Scatterplot of productivity VS quality (p. 28) =====
<code python>
import pandas as pd
import matplotlib.pyplot as plt
scatter_plot = pd.read_csv('D:\Python\Libri\A_Casebook_for_a_First_Course_in_Statistics_and_Data_Analysis_Datasets\Data\Tab\\prdq.TAB', '\t')
productivity_Y = scatter_plot['Producti']
quality_X = scatter_plot['Quality']
plt.scatter(productivity_Y, quality_X, bins=20, colors='r')
ax = plt.gca()
ax.set_Xlabel('Assembly defects per 100 cars')
ax.set_Ylabel('Hours per vehicle')
ax.set_title('Scatter Plot of PRODUCTIVITY VS QUALITY')
plt.show()
</code>


===== Productivity versus quality in the assembly plant (p.29) =====

It worked the first time but now it doesn't work again. Maybe again a windows error?

<code python>
#1
import matplotlib.pyplot as plt
import pandas as pd
data_comparison = pd.read_csv('D:\Python\Libri\A_Casebook_for_a_First_Course_in_Statistics_and_Data_Analysis_Datasets\Data\Tab\prdq.TAB', '\t')
non_japanese = data_comparison.loc[data_comparison['QualNonJ']]
japanese = data_comparison.loc[data_comparison['QualJapn']]
plt.boxplot([non_japanese['Quality'],japanese['Quality']], labels= ['Non-japanese','Japanese'])
plt.show()

#2
import matplotlib.pyplot as plt
import pandas as pd
data_comparison = pd.read_csv('D:\Python\Libri\A_Casebook_for_a_First_Course_in_Statistics_and_Data_Analysis_Datasets\Data\Tab\prdq.TAB', '\t')
non_japanese = data_comparison.loc[data_comparison['ProdNonJ']]
japanese = data_comparison.loc[data_comparison['ProdJapn']]
plt.boxplot([non_japanese['Producti'],japanese['Producti']], labels= ['Non-japanese','Japanese'])
plt.show()
</code>