Question

Applying statistical analysis skills to real-world decision making is key in modern business and it can make a company to be ahead competitively. Even in today’s workplace, you can have an immediate competitive edge over other new employees, and even those with more experience, by applying statistical analysis skills. Chose any company that you have observed that it is not utilizing its data as your case study. Do some background research on the company? Write an essay (report) outlining some statistical data analysis that the company can use in its decision making. Relate how data analysis can be attained by using built-in R-programming packages or functions.

Answer 1

With the changing global landscape of how businesses are being conducted, i.e., we are no longer an economy of products and services. Today, data is the new oil. Proper analysis of data has the ability to draw out such meaningful insights that may go unnoticed otherwise. Statistical analysis plays an imperative role in driving today’s core enterprise business activities, from strategy to operations.

Unfortunately, a study conducted by Forbes Magazine stated that even today more than fifty percent companies do not used advanced statistical analysis to drive results. Why? Perhaps the data isn’t clear. It’s hard to measure to the volume of data being generated. The pace of change is accelerating faster than ever. Lack of skills causing a hindrance, for example a study conducted stated that there is a global shortage of 55% for the required skill set. The list goes on.

Statistical Analysis can be in the form of risk analytics, financial analytics, HR analytics, marketing and research analytics, supply chain analytics, etc. Any company can use statistical analysis right from the beginning to the end of any delivery cycle and it acts as a quintessential ingredient in every step of decision making.

There are various R packages for different stages in Data Analysis;

1. Pre-modelling Stage:

• Data visualization ggplot2.
• Data Transformation: plyr
• Missing Value imputation: MissForest
• Outlier Detection: Outliers
• Feature Selection: Features
• Dimension Reduction: FactoMineR

1. Modelling Stage:

• Continuous Regression: randomforest
• Ordinal Regression: RMiner
• Time Series: forecast
• Classification: Caret
• Clustering: RankCluster

1. Post Modelling Stage:

• General Model Validation: Comparison
• Regression validation: RegTest
• ClassificationValidation: BinomTools
• ROC Analysis: Time Roc

However there are some other libraries that help in data analysis;

1. Improve performance: Rcpp
2. Text Mining: tm
3. Work with web: XML
4. Report Results: shiny, R markdown
5. Database: RODBC, RMongo

In the era of social and mobile technology, customers, employees, suppliers and partners are in direct communication with each other. Hence there is a vast amount of data that is generated that can influence the decision making and their spending.

By leveraging the changing trends and effective use of Statistical Analysis, companies can delight not only their customers but also employees with every interaction.