Statistics for Business Analytics: Data Analysis with Excel

Learn to compute the mode in Excel for continuous data and apply a workaround for categorical data using countif, max, and index/match, including handling crashes and selecting data ranges.

Master the coefficient of variation, a relative variability measure defined as standard deviation over the mean times 100, with stock returns and burger juiciness examples in Excel.

Explore covariance as a measure of how two variables move together and compare it to correlation, noting direction and strength from a sample perspective.

Understand the standard error of the sample mean and how it differs from standard deviation, and see how it informs confidence intervals by comparing sample and population means.

Learn how the z score standardizes values by subtracting the mean and dividing by the standard deviation, enabling comparisons across different scales with real-life GPA and GMAT examples.

Apply confidence intervals for large samples using the standard error of the mean and the z value at a chosen level. Apple's 95% interval guides potential demand and decisions.

Develop confidence intervals for small samples using the t distribution and margin of error. See an Excel example estimating population values with bounds and the impact of sample size.

Explore degrees of freedom as the number of values that can vary under constraints, and why n minus one governs confidence intervals with the t distribution, sample mean and variance.

Encourage learners to complete the feedback form, share what's firing them up and what's missing, and fuel the halfway-through course toward the next half with greater energy.

Master hypothesis testing with an Amazon fashion returns study, learning p-values, null hypotheses, type I and II errors, publication bias, two-sample t tests, and one-tailed vs two-tailed decisions in Excel.

Understand the p value as the likelihood results are due to chance and how it guides rejecting or not rejecting the null hypothesis in hypothesis testing.

Analyze publication bias in statistics by comparing published and unpublished antidepressant trial data, revealing that positive results distort evidence and may mask weaker effectiveness and safety risks.

Test a hypothesis with unknown variance using the t distribution in a two-tailed t test. Compare a sample mean of 6.5 to a population mean of 6 using Excel.

Use a two-sample t test to compare means of groups and determine significance, illustrated with pizza consumption by horror versus romantic movie watchers, including variance checks and Welch's t test.

This lecture demonstrates performing linear regression with dummy variables in Excel, converting categorical fields into binary columns, avoiding the dummy variable trap, and interpreting coefficients for price.

This case study explains linearity bias in statistics and business analytics, showing how extrapolating a straight line with non-linear data yields flawed conclusions and limited validity to similar data.

Explore multilinear regression by extending linear regression with multiple predictors to explain diamond price, covering predictors such as color, clarity, and owner, and addressing issues like multicollinearity and overfitting.

Explore how to perform multilinear regression in Excel 16 using the data analysis toolpak, including creating dummy variables for color and clarity, interpreting coefficients, and assessing adjusted R square.

Assess regression accuracy using MAE, RMSE, and MAP to compare actual values with predictions; understand their interpretability, handling of outliers, and how to compute them in Excel.

Learn to assess a regression model in Excel by calculating mean absolute error, mean absolute percentage error, and root mean squared error from train and test data.