Data Mining Skill, MDS Program 1, MDS Program 3
Data Science in SAS ( Explore Predictive Modelling and Machine Learning)
0( 0 REVIEWS )
1 STUDENTS
Instructors
Data Science in SAS ( Explore Predictive Modelling and Machine Learning)
1 STUDENTS ENROLLED
Analytics is the use of:
Data, information technology, statistical analysis, quantitative methods, and mathematical computer-based models to help data scientists to gain improved insight about their business operations and make better, fact-based decisions
- Descriptive analytics
- uses data to understand past and present
- Predictive analytics
- analyzes past performance
- Prescriptive analytics
- uses optimization techniques
Course Curriculum
Session - 1 Introduction to Statistical Analysis | |||
What is Statistics? | 00:00:00 | ||
Basic Statistical Concepts in Business Analytics | 00:00:00 | ||
<ul> <li>Population</li> <li>Sample</li> <li>Variable</li> <li>Variable Types in Predictive Modeling Context</li> <li>Parameter</li> <li>Statistic</li> <li>Example Exercise</li> </ul> | |||
Statistical Analysis Methods | 00:00:00 | ||
<ul> <li>Descriptive Statistics</li> <li>Inferential Statistics</li> <li>Predictive Statistics</li> </ul> | |||
Solving a Problem Using Statistical Analysis | 00:00:00 | ||
<ul> <li>Setting Up Business Objective and Planning</li> <li>The Data Preparation</li> <li>Descriptive Analysis and Visualization</li> <li>Predictive Modeling</li> <li>Model Validation</li> <li>Model Implementation</li> </ul> | |||
An Example from the Real World: Credit Risk Life Cycle | 00:00:00 | ||
<ul> <li>Business Objective and Planning</li> <li>Data Preparation</li> <li>Descriptive Analysis and Visualization</li> <li>Predictive Modeling</li> <li>Model Validation</li> <li>Model Implementation</li> <li>Exercise</li> </ul> | |||
Session-2 Basic Descriptive Statistics and Reporting in SAS | |||
Rudimentary Forms of Data Analysis | 00:00:00 | ||
<ul> <li>Simply Print the Data</li> <li>Print and Various Options of Print in SAS</li> </ul> | |||
Summary Statistics | 00:00:00 | ||
<ul> <li>Central Tendencies</li> <li>Calculating Central Tendencies in SAS</li> <li>What Is Dispersion?</li> <li>Calculating Dispersion Using SAS</li> <li>Quantiles</li> <li>Calculating Quantiles Using SAS</li> <li>Box Plots</li> <li>Creating Boxplots Using SAS</li> </ul> | |||
Bivariate Analysis | 00:00:00 | ||
Session - 3 Data Exploration, Validation, and Data Sanitization | |||
Data Exploration Steps in a Statistical Data Analysis Life Cycle | 00:00:00 | ||
<ol> <li>Example: Contact Center Call Volumes</li> </ol> | |||
Need for Data Exploration and Validation | 00:00:00 | ||
Issues with the Real-World Data and How to Solve Them | 00:00:00 | ||
<ul> <li>Missing Values</li> <li>The Outliers</li> <li>Manual Inspection of the Dataset Is Not a Practical Solution</li> <li>Removing Records Is Not Always the Right Way</li> </ul> | |||
Understanding and Preparing the Data | 00:00:00 | ||
<ul> <li>Data Exploration</li> <li>Data Validation</li> <li>Data Cleaning</li> </ul> | |||
Sesssion - 4 Testing of Hypothesis Testing: An Analogy from Everyday Life | |||
What Is the Process of Testing a Hypothesis? | 00:00:00 | ||
<ul> <li>State the Null Hypothesis on the Population: Null Hypothesis (H0)</li> <li>Alternate Hypothesis (H1)</li> <li>Sampling Distribution</li> <li>Central Limit Theorem</li> <li>Test Statistic</li> <li>Inference</li> <li>Critical Values and Critical Region</li> <li>Confidence Interval</li> </ul> | |||
Tests | 00:00:00 | ||
<ul> <li>T-test for Mean</li> <li>Case Study: Testing for the Mean in SAS</li> <li>Other Test Examples</li> <li>Two-Tailed and Single-Tailed Tests</li> <li>Exercise</li> </ul> | |||
Session - 5 Correlations and Linear Regression | |||
What is Correlations? | 00:00:00 | ||
<ul> <li>Pearson’s Correlation Coefficient (r)</li> <li>Variance and Covariance</li> <li>Correlation Matrix</li> <li>Calculating Correlation Coefficient Using SAS</li> <li>Correlation Limits and Strength of Association</li> <li>Properties and Limitations of Correlation Coefficient (r)</li> <li>Some Examples on Limitations of Correlation</li> <li>Correlation vs. Causation</li> <li>Correlation Example</li> <li>Correlation Summary</li> </ul> | |||
Linear Regression | 00:00:00 | ||
<ul> <li>Correlation to Regression</li> <li>Estimation Example</li> </ul> | |||
Simple Linear Regression | 00:00:00 | ||
<ul> <li>Regression Line Fitting Using Least Squares</li> <li>The Beta Coefficients: Example 1</li> <li>How Good Is My Model?</li> </ul> | |||
When Linear Regression Can’t Be Applied | 00:00:00 | ||
Session- 6 Logistic Regression | |||
Predicting Ice-Cream Sales: Example | 00:00:00 | ||
Nonlinear Regression | 00:00:00 | ||
Logistic Regression | 00:00:00 | ||
Logistic Regression Using SAS | 00:00:00 | ||
SAS Logistic Regression Output Explanation | 00:00:00 | ||
<ul> <li>Output Part 1: Response Variable Summary</li> <li>Output Part 2: Model Fit Summary</li> <li>Output Part 3: Test for Regression Coefficients</li> <li>Output Part 4: The Beta Coefficients and Odds Ratio</li> <li>Output Part 5: Validation Statistics</li> </ul> | |||
Individual Impact of Independent Variables | 00:00:00 | ||
Goodness of Fit for Logistic Regression | 00:00:00 | ||
<ul> <li>Chi-square Test</li> <li>Concordance</li> </ul> | |||
Prediction Using Logistic Regression | 00:00:00 | ||
Multicollinearity in Logistic Regression | 00:00:00 | ||
<ul> <li>No VIF Option in PROC LOGISTIC</li> </ul> | |||
Logistic Regression Final Check List | 00:00:00 | ||
Loan Default Prediction Case Study | 00:00:00 | ||
<ul> <li>Background and Problem Statement</li> <li>Objective</li> <li>Data Set</li> <li>Model Building</li> <li>Final Model Equation and Prediction Using the Model</li> </ul> | |||
Session - 7 Time Series Analysis and Forecasting | |||
What Is a Time-Series Process? | 00:00:00 | ||
Main Phases of Time-Series Analysis | 00:00:00 | ||
Modeling Methodologies | 00:00:00 | ||
Jenkins Approach | 00:00:00 | ||
<ul> <li>What Is ARIMA?</li> <li>The AR Process</li> <li>The MA Process</li> <li>ARMA Process</li> </ul> | |||
Understanding ARIMA Using an Eyesight Measurement Analogy | 00:00:00 | ||
Steps in the Box–Jenkins Approach | 00:00:00 | ||
<ul> <li>Step 1: Testing Whether the Time Series Is Stationary</li> <li>Step 2: Identifying the Model</li> <li>Step 3: Estimating the Parameters</li> <li>Step 4: Forecasting Using the Model</li> <li>Case Study: Time-Series Forecasting Using the SAS Example</li> <li>Checking the Model Accuracy</li> </ul> | |||
Session - 8 Cluster Analysis | |||
What is cluster analysis | 00:00:00 | ||
Customer segmentation introduction | 00:00:00 | ||
What is distance matrix | 00:00:00 | ||
K-Means clustering algorithm | 00:00:00 | ||
Super market customer segmentation case study | 00:00:00 |