Minitab Mastery: Statistical Analysis and Data Visualization

Master predictive modeling in Minitab with advanced regression, decision trees, and descriptive statistics to analyze business data across sectors.

Assess the significance of predictors—close price, shares, and trades—on total turnover, with r-squared near 92% and a significant y intercept. The spread shows limited impact.

Explore how the open/close spread serves as a dummy predictor in a logistic regression to explain total turnover, using close price, shares, and trades; interpret coefficients and intercept.

Explore scatter plots and a regression model to show that shares and trades drive total turnover, while close price has little effect; analyze correlations and groups to validate the trends.

Explore multinomial and polynomial regression, derive quadratic equations y equals a x squared plus b x plus c, and interpret slopes and intercepts for predicting turnover from average trade size.

The lecture demonstrates quadratic regression with a single predictor, average trade size, to model average turnover using the equation y = a x^2 + b x + c.

Compare quadratic and linear regression via scatter plots and p values to assess model fit. Linear regression offers a better fit, with high price not predicting total turnover.

Learn to build decision trees for regression and classification using Minitab, handling categorical and numeric attributes to classify data and reveal patterns.

Explore decision trees, including root and leaf nodes, parent and child nodes, and binary and m-ary trees, highlighting interpretation, data types, and heuristic training as building blocks for ensemble methods.

Master CART regression interpretation by comparing trees, assessing R-squared stability across terminal nodes, and evaluating variable importance and potential biases in race and arrests.

Explore how cart regression builds a decision tree to predict sodium–potassium outcomes, interpret root and terminal nodes with R-squared, and reveal limitations when drug type dominates.

Use Minitab to generate Pearson correlation outputs for high-low spread and close-open spread, and for open, high, and low price returns, revealing open–low 0.702 and open–high 0.561 correlations.

Analyze spread and price data through observations and correlations, interpret open, high, low, and returns relationships, and verify results with scatter plots and regression to inform trading strategies.

Interpret high and low correlations and their implications for a pair trading strategy, then learn to generate and arrange scatter plots in Minitab Graph Builder, noting no regression line option.

Explore how city housing indices relate to the all-india index using scatter plots and regression outputs, including r-squared, p-values, and line of fit.

Master descriptive statistics in Minitab by computing means, standard deviation, t-tests, and exploring skewness and kurtosis, using real mutual fund return data and visualizations.

Analyze descriptive statistics for customer complaints and resting heart rate in Minitab, examining skewness, mean, and standard deviation. Compare before and after conditions to draw practical conclusions from the data.

Compare before and after resting heart rate using mean, median, and standard deviation in Minitab, and assess data quality and descriptive statistics for loan applicants to support predictive modeling.

Sort randomized data in Minitab by gender and ethnicity to reveal BMI patterns, storing results in new columns or a new worksheet.

Create pie charts in Minitab to show ethnicity and subject distributions, using unique values or table counts, then add slice labels and titles and save as jpeg, png, or tiff.

Compute the mean and standard deviation for a discrete random variable using a binomial distribution in Minitab from a data set of 15 events counting married people.

Explore one-way analysis of variance in Minitab to test differences between means of open and close prices, with equal variances, 95% confidence intervals, and interpretation of p-values and f-values.

Develop and interpret a regression model in Excel, with y = mx + c; identify m = 0.1354 and intercept = -0.0020, and assess the t-statistic and p-value for significance.

Explore predictive modeling, including regression and correlation, and implement them using Minitab and Excel for data analysis.

Explore descriptive statistics with Minitab, comparing mean, standard deviation, and skewness across finance, medical, and energy datasets. Analyze daily customer complaints data to interpret variability and distribution.

Explore basic correlation techniques, including positive, negative, and zero correlation, the correlation coefficient r, and how to interpret results to predict relationships using Minitab.

Explore correlation analysis in minitab, compare Karl Pearson's and Spearman's rho, create a store matrix, and explain why five variables yield a 4x4 matrix due to the unitary matrix.

Generate scatter plots with regression to visualize correlations between two variables, identify positive or negative relationships, and use a trend line to visually validate the data.

Demonstrate scatter plots with regression lines in Excel to reveal positive correlations across variable pairs, including heartbeat data, and interpret correlation values like 71.6% to validate trends in predictive modeling.

Analyze descriptive statistics for the machine setting as the independent variable, including min, max, range, mean, and standard deviation, and show energy rises by 0.216 per unit.

Explain r squared as the percentage of change in dependent variable explained by independent variable, and interpret regression equation y = mx + c with t and p values.

Generate basic statistics to predict stiffness from density and temperature using regression, compare models with and without temperature, and assess significance and model fit with predicted values and scatter plots.

Explore multicollinearity, its impact on regression when r-squared rises with more predictors, and how simple linear regressions help isolate effects for predictive modeling.

Interpret regression results to identify significant predictors using p-values, including insolation and east, north, and south directions, while noting time of day as insignificant; R-squared 89.88% shows a good fit.

Explore how to interpret and implement regression on a dataset with groups, using dummy variables for categorical predictors and predicting sales from client count and years in a Minitab workflow.

Analyze regression outputs from tabulated data, including R square, t values, and p values, to identify significant variables, and use scatter plots to confirm correlations between sales, clients, and years.

this lecture interprets a regression analysis of business metrics, highlighting an r-squared of 81.69% and how sales rise with client count and years in business across three groups.

Analyze ad effectiveness for a cereal using logistic regression. Use income as a predictor and children and ad viewing as categories to form four regression equations for buying decisions.

Explore logistic regression to determine how age, education, debt, and savings influence income for credit card users and non-users, using a live dataset and regression equations in Minitab.

Explains regression results on credit card data, showing age and education as significant income predictors; debt is insignificant, with income rising 4,206 per education year and 2,843 per age year.

Learn to perform basic predictive modeling in Excel using the data analysis toolpak, including ANOVA, t-test, correlation, descriptive statistics, and simple linear regression, with practical dataset examples.

Explore descriptive statistics in Excel through data analysis, setting input ranges with labels, and viewing means, standard error, deviation, and confidence intervals at 90% and 95%.

Learn how to perform single-factor ANOVA in Microsoft Excel using data analysis toolpak with two sets of scores, including setting alpha to 0.05 and interpreting the F and p values.

Build a regression model in minitab to predict total turnover from close price, number of shares, and number of trades, then interpret t-values, p-values, durbin-watson, anova, and the regression equation.

Analyze a multiple regression of Tech Mahindra data, revealing an r square of 96% with predictors—close price, number of shares, and number of trades—that explain most of the total turnover.

Compare linear and quadratic regression models using advanced Minitab techniques to predict total turnover from shares traded and number of trades, with regression outputs, scatter plots, and parabolic insights.

In Minitab Mastery, analyze waist versus weight using scatter plots and regression. Derive and interpret the regression equation, residuals, and ANOVA to predict weight from waist.

Frame null and alternate hypotheses as mutually exclusive for a two-tailed test, then use p-values and 95% confidence intervals to interpret data in Minitab.

Assess two hypothesis tests in Minitab: the mean equals 1001 with p = 0.979, and the defective proportion stays at or below 2% with p = 0.001.