Pre- or corequisite(s): ST 625
The course examines methods for analyzing financial time series. In most times series, observations from different times periods are correlated, which require a treatment that is different from usual regression analysis methods. The course reviews regression and covers smoothing and decomposition time-series models, Box-Jenkins analysis and its extensions, and other modeling techniques commonly used in business and finance, such as quantile estimation and value at risk.The course utilizes the R statistics platform.
Prerequisite(s): ST 625 or ST 635 or Instructor Permission
This course will introduce participants to the most recent data mining techniques, with an emphasis on: getting a general understanding of how the method works, understanding how to perform the analysis using suitable available software, understanding how to interpret the results in a business research context, and developing the capacity to critically read published research articles which make use of the technique. Contents may vary according to the interest of participants.
Topics may include decision trees, neural nets,self organizing (Kohonen) maps, multiple adaptive regression splines (MARS), genetic algorithms, association (also known as market basket) analysis, web mining and text mining, and social networks anaylsis..
Prerequisite(s): BF 501 or PPF 501 or GR 521
This course provides students with an in depth coverage of regression methods and an introduction to the anaylsis of time series data. Topics include simple and multiple linear regression techniques, the use of transformations such as squares and logarithms, interactions, heteroscedasticity and multicollinearity. Issues with outlying and influential observations are also covered.The art and science of model building are demonstrated with the help of practical applications from business and finance. An introduction to autocorrelation and the modeling of time series is provided. The course utilizes statistical packages such as SAS and SPSS.
Prerequisite(s): ST 625 or Instructor Permission
The course focuses on modeling situations dependent on multiple variables, as commonly found in many business applications. The issues addressed include: How do we predict who is more likely to respond to a direct mail offer? How can we identify important segments in our customer base? How do we summarize large sets of variables? A central objective of the course is for participants to be able to determine the appropriate multivariate methodology based on the anaylsis objectives and available data, carry out the analysis and interpret the results. Typical topics include logistic regression, cluster analysis, factor analysis, decision trees and other multivariate topics as time permits. The course utilizes statistical packages such as SPS and SPSS.
Prerequisite(s): ST 625 and at least one other Business Analytics concentration course
This course provides an opportunity for students to apply quantitative and data analysis skills in a live employment environment, serving as a quantitative analyst. With help from the internship coordinator, students identify a suitable internship and meet regularly with the internship coordinator. Students prepare a paper that discusses the internship experience and demonstrates at least one specific case analyzed during the internship period.