During this 5-day bootcamp on applied statistical analysis, attendees will be introduced to the theory and application of numerous statistical methods. Each instruction block will cover the rationale, limitations, assumptions, and sensitivities of each method, with rules-of-thumb for ensuring optimatization of resource allocation. Example applications for each method will be covered using empirical public domain data, with comparison of results for the methods and data considered.
Attendees will receive a licensed Enterprise version of the Explorer package (windows only), and the 388-page User's Guide. Computer labs will be performed after each topic (subtopic) is covered, so that implementation or interpretation questions can be answered during the example runs.
INTRODUCTION
- Scales: Nominal, Ordinal, Dichotomous(Binary), Discrete, Continuous
- Feature types (Continuous, Nominal, Binary, Text)
- General Default Preferences
- Color and Graphics Defaults
- Changing Feature Types
SUMMARY STATISTICS
- Data Arrangement
- Normal Distribution
- Central Tendency and Dispersion
- Histograms, X-Y Scatter plots, Matrix plots
- Skewness, Kurtosis
- Normality Tests
- Heteroscedasticity Tests
LABELS, TRANSFORMS, AND FILTERING
- Editing labels, mathematical transformation
- Filtering records
TRANSFORMATIONS
- Z-Scores from Standard Normal Distribution
- Log
- Quantile
- Rank
- Percentile
- van der Waerden Scores
- Nominal-to-Binary
- Fuzzification
- Fast Wavelet Transform (FWT)
- Root MUSIC
- Text to Categorical
INDEPENDENCE
- 2 Unrelated Samples
- k Unrelated Samples
- Equality Tests for 2 Independent Proportions
- Chi-Squared Contingency Tables
- Nominal (Categorical) Measurement Scale
- Two-way Contingency Tables
- Fisher's Exact Test
- Multiway Contingency Tables
- Exact Tests for Multiway Tables
- Related Samples
ASSOCIATION
- Covariance
- Parametric Correlation: Pearson Product Moment
- Force plots
- Non-parametric Correlation: Spearman Rank
- Multivariate Forms of Covariance and Correlation
- Euclidean Distance
- Matrix Formulation of Covariance and Correlation
- Association Rules (Market Basket Analysis)
DEPENDENCY
- Linear Regression: Single Predictor (and Diagnostics)
- Multiple Linear Regression (and Diagnostics)
- Residuals [standardized, Studentized, Jacknife(deletion), leverages, Cook's D, DFFITS, DFBETAS]
- Multicollinearity (Variance inflation factors),
- Multivariate Linear Regression
- Binary Logistic Regression (Diagnostics)
- Polytomous Logistic Regression
- Additive and Multiplicative Poisson Regression (Diagnostics)
- Non-Linear Poisson Regression
SURVIVAL ANALYSIS
- Kaplan-Meier Analysis and Logrank Test
- Cox Proportional Hazards Regression
MONTE CARLO UNCERTAINTY ANALYSIS
- Empirical Cumulative Distribution Fitting
- Probability Distributions
- Simulation of Correlated Features
- Parameter Specification for Manual Input
- Correlation Matrices
MONTE CARLO COST ANALYSIS
- Effect of correlation
- Cost risk analysis
- Project: Construction of new 3-story building:
-Design costs & schedule
-Earthworks costs & schedule
-Foundation costs & schedule
-Structure (floors,pillars,roofing) costs & schedule
-Envelope (walls,windows,external doors) costs & schedule
-Services(plumbing,electrical,cabling) & Finishes (partitions,decorations) & schedule
-Site cleaup & Landscaping costs & schedule
