Repository for DS 200 Research Methods

Three datasets are plotted in the given assignment.

1. The first dataset plotted as a box plot is the Mean temperature of India between 1901-2012 annually as well as over seasons. The plot has 4 box plots for each such season. It shows the minimum, maximum, 25 quartile, median and 75 quartile values within each such box plot. Explaining further, the first box plot shows that the median temperature of Jan-Feb has been ~19 degrees Celsius and the 25 quartile value has been just over 18 degrees Celsius.

Observations-

1. The median temperature in Jan-Feb has been ~19 deg C with minimum having been just over 18 deg C and maximum just over 20 deg C
2. The median temperature in Mar-May has been 26 deg C with minimum having been just over 25 deg C and maximum just over 27 deg C
3. The median temperature in Jun-Sep has been ~27 deg C with minimum having been just over 26.5 deg C and maximum just over 27.5 deg C
4. The median temperature in Oct-Dec has been ~22 deg C with minimum having been just over 21.5 deg C and maximum just over 22 deg C

Dataset source- https://data.gov.in/catalog/annual-and-seasonal-mean-temperature-india Code citations- a) https://pandas.pydata.org/ b) https://github.com/simmhan/probable-bassoon/blob/master/plot-scripts/BoxPlot.py

2. The second dataset plotted as a scatter plot is the comparison between GDP per capita and expenditure on R&D per capita for a list of 29 countries. The data points compare the expenditure on R&D and the GDP per capita for these countries and provide us an inside look at what percentage of expenditure each country makes from its per capita GDP. Each data point corresponds to a particular country. The data points haven't been labelled and thus the dataset needs to be checked to infer this. The code will be updated to reflect this issue.

Dataset source- https://data.gov.in/resources/rd-expenditure-capita-and-percentage-gdp-selected-countries-2009-us Code citations- a) https://pandas.pydata.org/ b) https://github.com/simmhan/probable-bassoon/blob/master/plot-scripts/ScatterPlot.py

3. The third dataset plotted as a line plot is the pattern of overall value of mineral production data in India from 2000-2016 and gives us an insight into how the production value has increased or stayed constant over the years and this is a very insightful plot in the sense that it allows us to predict further production values based on the given plot and therefore prepare adequate resources. Each datapoint is marked on the line with a box.

Observations- The value of minerals produced in the country has seen a steady increase from 2000 to 2016, reaching a peak production in 2013-14. We also there was a huge spike in production from 2007-08 to 2008-09 with the values going up from just under 1.1x1e9 to 1.7x1e9 and then again from 2009-10 to 2010-11 going from just under 2x1e9 to 2.5x1e9 value

Dataset source- https://data.gov.in/sites/default/files/datafile/Table_15.2-All_India_MINING_1_1.csv Code citations- a) https://pandas.pydata.org/ b) https://github.com/simmhan/probable-bassoon/blob/master/plot-scripts/LinePlot.py