Practical Data Science and Statistical Learning

There are two files and one folder to practice data science and statistical learning step by step.

File StatisticalLearning.r

The first file, StatisticalLearning.r, contains five queations, which cover factor selection, initial analysis, statistical learning on both qualitative and quantitative data.

Question 1 (Introduce regsubsets function)

Demonstrate best subsets, forward selection and backwards elimination to identify a great subset of Auto and take in to my statistical model. Auto is a dataset on http://www-bcf.usc.edu/~gareth/ISL/Auto.data

Question 2 (Exploratory data analysis)

Get ideas how to do initial analysis using Boston dataset, which is also a dataset in ISLR.

How many columns?

What is the range of each quatitative variables?

Is there any relation between two variables?

How many data while given constraints?

Question 3 (Do statistical learning on binary outcomes)

Train Weekly dataset, which is also a dataset in ISLR, using Logistic regression, Linear Discriminant Analysis, Quadratic discriminant analysis, and K - nearest neighbor.

Question 4 (Write a function of linear regression without using 'lm')

Create a function of linear regression on binary outcomes

Question 5 (Do statistical learning on quantitative data)

Train, test, and predict sales, which is set as response among others factors of ads, using linear regression. ads is a dataset on http://www-bcf.usc.edu/~gareth/ISL/Advertising.csv

File NonlinearRegressions.r

The second file, NonlinearRegressions.r, contains five queations, which cover cross-validation, generalized additive models, classification tree, random forest, gradient boosted machines, and regularized generalized linear models on quantitative data.

Question 1 (Cross-validation for generalized linear models)

Compare errors using leave-one-out cross validation on different polynomial terms

Question 2 (Generalized additive models)

Do statistical learning on out-of-state tuition against other factors in College dataset using gam. College is a dataset in ISLR.

Question 3 (Do statistical learning on binary outcomes)

Perform cross-validation to choose the optimal number of cuts on Age, a factor in Wage dataset, and then find a fit regression to predict wage.

Question 4 (Classification tree and random forest)

In the lab, a classification tree was applied to the Carseats data set after converting Sales into a qualitative response variable. Now we will seek to predict Sales using regression trees and related approaches, treating the response as a quantitative variable.

Question 5 (Gradient boosted machines and regularized generalized linear models)

Use boosting (and bagging) to predict Salary in the Hitters, a dataset in ISLR. Aslo, compare the mean-squared errors of linear regression and ridge regression.

Folder ExploratoryDataAnalysis

The folder, ExploratoryDataAnalysis, has a rmd file, a pdf file, and a dataset folder. The pdf file was kintted by the rmd file. First, I discussed exploratory data analysis on the datasets provided for Bronx, Brooklyn, Manhattan, Queens, and Staten Island to visualize and make comparisons for residential building. Next, I did the exploratory data analysis on datasets provided nyt1.csv, nyt2.csv, and nyt3.csv to visualize some metrics and distributions over time.

Reference:

Gareth James, Daniela Witten, Trevor Hastie, and Robert Tibshirani. An Introduction to Statistical Learning with Applications in R. Springer, 2014. (1) This book is available free from the author's web site at http://www-bcf.usc.edu/~gareth/ISL/ISLR%20Fourth%20Printing.pdf (Links to an external site.)Links to an external site.

Trevor Hastie, Robert Tibshirani, and Jerome Friedman. The Elements of Statistical Learning: Data Mining, Inference, and Prediction. Second Edition., Springer (Tenth Printing) 2013. (2) This book is available free from the author's web site at http://statweb.stanford.edu/~tibs/ElemStatLearn/printings/ESLII_print10.pdf