SYST/OR 568. Applied Predictive Analytics - Mason Analytics MS

1. DataCamp. You need to sign up for it, using the invite link. You will need to use your @masonlive.gmu.edu email to get free access.

2. Stanford's Statistical Learning class.

3. ISLR Book

4. Piazza for discussing HW/project/midterm. You can sign up here

For each week I provide direct links to DataCamp courses and Statistical Learning videos.

Week of March 23

1. Datacamp: Working with Dates and Times in R

2. Resampling Methods (ISLR Chapter 5). Videos: a, b, c, d, e, HandsOn

Week of March 30

1. Midterm due. Please submit 3 files to the BB. (a) First page with your signature acknowledging the honor code (b) PDF file with you solution (do not copy-paste large code chunks, just essential parts that you added on top of my scripts) (c) zip file with R scripts.

Week of April 6

1. Linear Model Selection and Regularization (ISLR Chapter 6). Videos: a, b, c, d and e, f, g, h, i, j, k, HandsOn

2. Project proposals Due. Proposal is to be half-a-page and to contain: (a) description of the data to be analysed. (b) What is the goal of analysis, e.g. predict y from x or understand relations between y and x. (c) Why this is important? (d) What methods you think you will use? You can do it in bullet-points.

Week of April 13

1. HW3 Due

2. Tree-based Methods (ISLR Chapter 8). Videos: a, a2, b, c, d, HandsOn

3. Datacamp: Machine Learning with Tree-Based Models in R

Week of April 20

1. HW4 Due (ISLR: Ch 8, Ex. 4, 7, 8, 11)

2. Unsupervised Learning (ISLR Chapter 10). Videos: Unsupervised Learning (Chapter 10) a, b, c, d, e, HandsOn

3. Datacamp: Unsupervised Learning in R

Week of April 26

1. Project intermediate reports are Due

2. Life session discussing the projects

Week of April 27

1. HW5 Due (ISLR: Ch 10, Ex. 3, 10)

2. Optimization (I will post my videos on YouTube and will post links here when done)

3. Datacamp: Communicating with Data in the Tidyverse (Ch 3-4). Plus you can use rmarkdown materials on the official website

Week of May 4

1. Final Project Presentations Due. Delivered as html file generated from R Markdown. Your HTML pages will be published on the course page and will be publicly available.

Introduces predictive analytics with applications in engineering, business, finance,health care, and social economic areas. Topics include time series and cross-sectional data processing, data visualization, correlation, linear and multiple regressions, classification and clustering, time series decomposition, factor models and causal models, predictive modeling performance analysis, and case study. Provides a foundation of basic theory and methodology with applied examples to analyze large engineering, social, and econometric data for predictive decision making. Hands-on experiments with R will be emphasized.

Syllabus

Place and Time

• Fridays 4:30 pm - 7:10pm at Sandbridge Hall 107 (Jan 21, 2020 - May 13, 2020)

Grades

Grade composition: Grade based entirely on participation in class, homework assignments, in-class midterm and final project.

Textbooks

• Diez, Barr and Cetinkaya-Rundel OpenIntro Statistics, OpenIntro, 2015

• James, Witten, Hastie and Tibshirani, An Introduction to Statistical Learning with Applications in R, Springer, 2009.

• Kuhn and Johnson, Applied Predictive Modeling, Springer, 2013.

• Hyndman and Athanasopoulos, Forecasting: Principles and Practice, OTexts, 2013.

Links

• The State of Data Science and Machine Learning

• Model Interpretability

• Data Sceince Survey

• J. M. Keynes's Investment Performance: A Note

• Bayes Search

• How Y'all (NY Times)

• Monte Hall

• EPL

• Data Cleaning

• Airbnb (Random Forest)

• Facebook (Decision trees and logistic regrsssion)

• Youtube (deep learning)

• Uber (time series)

• Exploratory Data Analysis

• Overfitting

• Election 2016

Deep Learning

• Course by Jeremy Howard on coding aspects of deep learning

• Short introduction into deep learning

• Architectures

• Keras

• ConvNetJS

• Opt Visualization

• Momentum

• Backprop

Final Project Data

• Debby Kermer (data services): contact info

• Grocery Shopping

• Walmart

• Yelp

• Airbnb

• Zillow

• Uber

• FiveThirtyErigh

• UCE ML Repo (Lots of datasets along with descriptions of each)

• Knuggets (Lot of links to datasets relevant for data mining)

• Pew Research Center Survey datasets

• ExonData (The site has links to plenty of regional, state, and local economic data)

• 3stages (Searchable listing 363 Internet sites of Social Science data)

• Bureau of labor statistics

• data.gov (A repository for information collected by the federal government)

• Medicare datasets

• Bureau of Transportation Statistics

• United Nations

• Chicago (Urban Analytics)

• NYC Taxi Trips

• Virginia

Very biased and noisy course recommendations

Here are the courses that cover different aspects of data science

• Statistical modeling (STAT250, SYST664, OR719, STAT 554)

• Data management (AIT614)

• Optimization (OR604)

Notebooks (ISLR labs)

• Introduction

• Regression

• Classification

• CV

• Model Selection

• Nonlinear Models

• Trees

• SVM

• PCA

• All

Previous instances

• Fall 2016

• Spring 2017