Statistical Learning and Data Mining
Guide: Efficient asset allocation through statistical learning methods and comparison of techniques for the creation of an index tracking ETF (Exchange bought and sold fund) Datasets:
The datasets are selected from the internet site of the publication " Statistics and Info Analysis to get Financial EngineeringвЂќ by David Ruppert. The book can be mentioned among the references for this course. Both data sets chosen will be 1 . Stock_FX_Bond. csv
2 . Stock_FX_Bond_2004_to_2006. csv
The data contains the volumes of prints and altered closing rates for GM, F, UTX, CAT, MRK, PFE, MSFT, IBM, C and XOM. The data as well contains the amounts and adjusted closing prices for the S& G 500 index. The data arranged also includes treasury rates for different maturities and rates about corporate bonds as well as foreign currency rates for the period of 1987 to 2006.
1 . Ideal portfolios to get various amounts of Risk.
Regular investors look for attain maximum alpha beliefs (rates of return) in levels of risk they are more comfortable with. We can therefore at any level of risk, specify portfolios that generate maximum returns. Through this project, we aim to determine the structure of portfolios that defines this ideal objective.
Existing models just like CAPM, along with additional forms of regression will be used to compare with additional methods, not protected in the duration of the class to distinguish the better methods of stock portfolio creation. We all will use learning aids and models to predict the costs of go back and risk for each share that will allow all of us to build portfolios to suit requires. We will certainly carry out uncertainty analysis employing resampling tactics and make an attempt to use Bayesian methods too. The efficiency will be analyzed using future rates of return about these portfolios.
2 . Creation of an index tracking exchange traded tool
An ETF is an investment fund that is certainly comprised of various assets just like stocks, a genuine or...