  Class Details

Price: \$1,500﻿

Course Includes:

• Instructor-led classroom training and exercises
• Take-home code templates
• Course reference guides and cheat sheets for additional suport at home
• Case studies and interactive practice scenarios to help develop problem-solving skills
• Courseware books, notepads, pens, highlighters and other materials
• Full breakfast with variety of bagels, fruits, yogurt, doughnuts and juice
• Tea, coffee, and soda available all day
• Freshly baked cookies every afternoon - * only at participating locations

This 2-day course teaches data scientists and analysts how to forecast trends based on multiplle variable and factors. Students will learn how to forecast economic trends, utilization rates, customer demand, as well as a variety of other activities.

Course Outline

Module 1: Introduction to Regression and Time-Series Analysis

• Lesson 1: Commerical applications of regression and time-series analysis
• Lesson 2: Linear relationships: slope, y-intercept, variable interactions
• Lesson 3: Variance and standard deviation
• Lesson 4: Covariance and correlation
• Lesson 5: Normal distribution and bell curves

Module 2: Evaluating Your Model

• Lesson 1: Distribution of errors: Q-Q plot, heteroscedasticity
• Lesson 2: Multivariate regression
• Lesson 3: R- and adjusted R-
• Lesson 4: p-values and the t-test
• Lesson 5: F-test and F-distribution

Module 3: Identifying the Most Important Variables

• Lesson 1: Multicollinearity test
• Lesson 2: Heteroscedasticity test
• Lesson 3: Model selection: Akaike Information Criterion
• Lesson 4: Polynomial regression
• Lesson 5: Confidence intervals

Module 4: Time-Series Analysis Seasonlity

• Lesson 1: Moving averages
• Lesson 2: Seasonality detection: autocorrelation
• Lesson 3: Seasonality: additive vs. multiplicative
• Lesson 4: Decomposing seasonal data: trend, level and seasonality
• Lesson 5: Multiplicative Holt-Winters exponential smoothing
• Lesson 6: Forecasting seasonal trends
• Lesson 7: LOcal regrESSion: LOESS

Objectives

At the conclusion of this course, participants will be able to do the following:

• Build single and multivariate regression models
• Assess statistical significance and validate models for explanatory power and bias
• Use time-series models to identify seasonality patterns and create forecasts for cyclical data