Data Analytics with R, Python and SQL

Data analytics example. Demonstrates how data analytics plays out .

What does it take to interpret a data set. Big picture of the given data. What business may want to know, looking at the data. Ultimately what information can we gain out of analyzing the data. Uses the UK dataset as before for demonstrating the process of interpretation.

Using Statistics and Mathematics to gain better understanding of the data. Gain knowledge of the data and is key to start analyzing the data. Gives 30,000 foot view of the data.

People across the board prefers visuals, and so look for ways to visualize the data. Charts provide information more readily by looking at the charts, graphs, and visuals. Within a simple visual like the ones presented in this section enables the analyst a picture of how/what the data is saying.

Widely used data analysis programming languages R, and Python. Simple, easy to read language syntax and  abundant libraries to use. All you need know how to call the commands in the libraries or packages with the IDE, integrated development environment.

Implement typical scenarios in data analytics. Getting familiar with RStudio.

After the loading the data into R, may be modified, manipulated, and updated based on analytics need, and as determined by the data analyst, can be saved to a file for further use by another program and/or programmer. Continued from the previous section, and uses the same R script.

Working environment is a starting point and is demonstrated to get started with R and Python with RStudio as IDE (Integrated Development Environment). Download and install R and RStudio platform for coding.

Get a high level picture of the data from the summary function. This is the first step in the analysis. Uses Financial data as an example. Find the references for the data in the resources.

Observe the data closely. See for any missing data, null values, or NAs within the dataset. Remove the Nulls, and NAs.

Continue to learn data observation using filtering to slice and dice the data. Uses the same R script as last section to wrap up the clearing the Nulls, and/or NAs.

R Markdown to facilitate adding extensive commentary on the code, and when executed the comments are added to the output along with results from the code, making it powerful enough and ready to publish the work. 

Statistical methods defined and used to explore data to get relevant information on the data. This enables the analyst to pursue organize and arrange further analyze and dig deeper into the data. Demonstrates the use and implementing the methods to practical dataset.

Demonstrate ggplot and boxplot to get the Box/Whisker plots to get an understanding of the data in terms of basic statistics and gain information on the overall distribution of the data.

Significance of Box Plots. Demonstrate with example on how to use these plots to interpret or to describe the given data.

•Correlation coefficient (R) indicates the extent to which two variables follow the variation

Three correlation coefficients are popularly used – Pearson coefficient for continuous variables, only for linear relationship; and Spearman coefficient for continuous and or ordinal variables; only monotonic relationship; Kendall coefficient – like Spearman, but insensitive to error, and p values are accurate for small samples

•Linear relationship: variables move in the same direction at a consistent rate

•Monotonic relationship: changes happen in the same relative direction, but at an inconsistent rate

–Coefficient of variation or Variance is expressed as R2 (also called Coefficient of Determination) – which explains the variation

•Visualize relationships between the qualitative variables

•Splits the data based on the input data supplied