
learn how regression analysis explains variation in a dependent variable using multiple independent variables, predict outcomes, and address issues like multi-collinearity and qualitative variables, with a practical Excel walkthrough.
Analyze regression results to understand how variables such as sales culture, debt, capex, and the economy influence future sales, using coefficients, standard errors, significance, and R-squared to gauge predictive accuracy.
Evaluate commercial databases carefully to avoid black box data and governance issues, weighing manual versus automatic data entry, and consider Excel or your own data sets with appropriate analysis tools.
Identify and correct data errors in large data sets by dropping illogical values, flagging beyond three standard deviations, and applying mean, median, and standard deviation checks, Bedford's and Benford's laws.
Explore business intelligence fundamentals of data analysis, including an overview, univariate analysis for common business questions, regression analysis for forecasting, and advanced regression techniques, with a preview of next course.
Explore interpreting results from big data analytics and apply insights to your company, covering univariate analysis, regressions, economic significance, and project limitations.
Explore univariate analysis by examining means, medians, and percentiles over time; analyze pre- and post-change sales to understand significant impacts and driving factors.
Use multivariate regression to identify drivers of sales, evaluating variables like time with customers, sales commission, and marketing, via Excel's data analysis tool.
This course is broken up into four modules.
The first module will prepare participants to begin business intelligence projects at their own firm. The focus of the course is a hands-on approach to gathering and cleaning data. After taking this course, participants will be ready to create their own databases or oversee the creation of databases for their firm. The focus in this course is on “Big Data” datasets containing anywhere from tens of thousands to millions of observations. While the tools used are applicable for smaller datasets of a few hundred data points, the focus is on larger datasets. The course also helps participants with no experience in building datasets to start from scratch. Finally, the course is excellent for users of Salesforce, Tableau, Oracle, IBM, and other BI software packages since it helps viewers see through the “black box” to the underlying mechanics of Business Intelligence practices.
The second module will prepare participants to begin business intelligence projects at their own firm. The focus of the course is a hands-on approach to structuring data including generating new variables based on comparative and relative metrics. The structuring of these variables will be done in Excel, SAS, and Stata to give viewers a sense of familiarity with a variety of different software package structures. The focus in this course will be on financial data though the techniques are also applicable to more general forms of data like that used in marketing or management analyses.
The third module will prepare participants to begin running data analysis on databases. Both univariate and multivariate analysis will be covered with a particular focus on regression analysis. Regression analysis will be done in Excel, SAS, and Stata to give viewers a sense of familiarity with a variety of different software package structures. The focus in this course will be on financial data though the techniques are also applicable to more general forms of data like that used in marketing or management analyses.
The fourth and final module will prepare participants to review, analyze, and make decisions based on results from business intelligence projects. The course will cover reading and interpreting regression analysis. The course will also give participants the skills to critically analyze and identify potential limitations on analysis. The course will also cover predicting changes in business outcomes based on analysis and identifying the level of certainty or confidence around those predictions. This paves the way for future detailed courses in predictive analytics.