1. Please log in to Blackboard at (https://blackboard.pace.edu/webapps/portal/execute/tabs/tabAction?tab_tab_group_id=_2_1)

2. go to courses and click on the course, “ECO-240-21498.201720: Quantitative Anlys & Forcastng SPRING 2017 21498”

3. On the left hand side tab click on “assignments”

4. Please do “Assignment #1 due Thursday February 2 @ Midnight”

I have also attached on the additional materials: a document with the same instructions from blackboard and the syllabus for your access to MINITAB. I have highlighted in yellow on the “syllabus” document all the instructions to access the MINITAB software.
Please make sure to read the syllabus and follow the instructions.

Economics 240: Quantitative Analysis and Forecasting
Dr. Walter Morris
E-mail wmorris@pace.edu
Phone 914-773-3308
Office Hours (Choate 54): Tuesday 4-5PM, Wed 2-5PM, Thursday 8:30-9:30AM, and by appointment.

Textbook, Software, and Data Bases

(1) John E. Hanke and Dean W. Wichern, Business Forecasting, Pearson/Prentice Hall, Ninth Edition, 2009. (REQUIRED)

(2) There are three (3)options to access MINITAB (REQUIRED).

a. The MINITAB statistical/forecasting computer program is available free at a Pace Computer Lab located at the Pleasantville or NY campuses.

b. It is also available on the Virtual Pace Network (VPN)—see Assignment #1 for downloading instructions.

c. Also, for a six month period a student version for your PC can be ordered separately on-line directly from MINITAB (for a fee of approximately $30—it can be downloaded directly from MINITAB).

(3) The ECONOMAGIC data base which is supplied on the Pace Network and free.

Course Objectives

This course is designed to give students a detailed understanding of the mathematical methodologies associated with economic and financial forecasting. Emphasis is placed on five forms of forecasting: i.e., Regression Analysis, Exponential Smoothing, Time Series Decomposition, Autoregressive Integrated Moving Averaging (ARIMA) models, and Data Mining. Students will also be required to demonstrate competence using the statistical/econometric forecasting programs residing in MINITAB. There is also a brief introduction to the Data Mining computer software programs residing in SAS Enterprise Miner. Data bases, in particular ECONMAGIC, are also explored in detail.

At the conclusion of the course students are expected to:

1. Understand forecasting theories as they relate to current economic/financial issues.
2. To evaluate economic issues from a variety of different forecasting perspectives.
3. To critically evaluate different research methodologies with regards to current forecasting problems.
4. To demonstrate critical analytical and thinking skills as they relate to forecasting issues.
5. Be able to evaluate forecasting issues from both a global as well as a national perspective.
6. To demonstrate competence in several econometric/forecasting computer packages such as MINITAB, SAS, SAS Enterprise Miner, and XLMiner.
7. To demonstrate competence in handling data bases such as ECONOMAGIC as applied to forecasting issues.

As partial fulfillment towards receiving credit for the course, students must sit for one mid-term examination worth 30 points. In addition, subject to the instructors’ approval, students are required to submit a 10-15 page research paper that demonstrates to the instructor a measure of competence in the forecasting area. The topic of the paper should be in a specific area of forecasting and it is worth an additional 30 points. Homework assignments located in the “Assignments’ folder in Blackboard count for 40% of the final mark. All assignments must be submitted as a WORD document. You may cut and paste results from the MINITAB output but it must be pasted into a WORD document!

Students are expected to attend and participate on-line. Several homework’s are assigned, required, and reviewed and are an integral part of the course. The homework is found under the ‘ASSIGNMENTS’ menu located in your BLACKBOARD account. All homework must be submitted as a WORD document to your folder on the date and time specified in the ‘Assignment’ folder. It cannot be emphasized enough that late submissions will not be allowed.

The instructor will use ECHO360 throughout the semester. Students are strongly advised that this course requires an intensive amount of preparation and it is to their advantage to carefully review these documents. The first document is titled ‘Introduction to MINITAB’ and you may wish to review it at your leisure. I will also use ECHO360 to highlight other issues pertaining to forecasting. The instructor will use the Discussion Board to answer questions related to the chapter. However, I will not directly answer questions assigned for homework

SUMMARY OF COURSE REQUIREMENTS

Mid-term Examination 30%
Homework Assignments 40%
Final Paper 30%

The Course Outline is an overview of selected topics which the instructor chooses from on a semester by semester basis. A significant portion of the course is tailored towards Topics 1, 2, 3, and 4 and depending on the interests of the class, Topics 5, 6, and/or 7 can be included.

Course Outline

Topic 1: Statistical Review and an Introduction to Forecasting
Chapters 1-3

A brief review of the statistical concepts and graphical techniques used in class. An exploration of data patterns and overview of forecasting techniques. Forecasting through the use of regression analysis is developed. A brief introduction to MINITAB and ECONOMAGIC is covered in this section and data sets are supplied.

Topic 2: Statistical Review and an Introduction to Regression Analysis
Chapter 6

A brief review of the statistical concepts used in class. The important univariate statistics are the t-test and the F-test (i.e., ANOVA). Bi-variate regression is introduced by estimating coefficients using Ordinary Least Squares. Hypothesis testing, confidence intervals, p-values, and goodness of fit measures are reviewed. Forecasting through the use of regression analysis is developed. Functional form or non linearity in the variables is covered. Advanced demonstrations of MINITAB and ECONOMAGIC.

Topic 3: Multiple Regressions and Forecasting
Chapters 7

The multiple regression model is estimated and interpreted. Multicollinearity is addressed by investigating its causes, consequences, tests-to-detect, and correction procedures. Dichotomous or Dummy Variables are examined. Point and interval forecasts are developed as are ex post and ex anti forecasting procedures. Forecasts are evaluated using conventional statistical tests (i.e., root mean square error, Theil’s inequality coefficient, etc.) Several economic/financial applications with regard to forecasting are introduced. Advanced demonstrations of MINITAB and ECONOMAGIC.

Topic 4: Time Series Data and Associated Forecasting Problems in Regression Analysis
Chapter 8

Single equation regression problems associated with time series data such as autocorrelation and heteroscedasticity are covered in detail. The approach here is to first identify the cause of the problem, then discuss the consequences, develop a test statistic to detect the problem, and finally come up with a correction procedure to remedy it. Point and interval forecasts are developed as are ex post and ex anti forecasting procedures. Forecasts are evaluated using conventional statistical tests (i.e., root mean square error, Theil’s inequality coefficient, etc.) Several economic/financial applications with regard to forecasting are introduced. Advanced demonstrations of MINITAB and ECONOMAGIC.

Topic 5: Introduction to Time Series Models
Chapters 4-5

Moving Averages and Exponential Smoothing methods are introduced. Decomposing time series data into trends, cycles, and seasonality’s are discussed. Advanced demonstrations of MINITAB and ECONOMAGIC are covered.

Topic 6: Time Series Models
Chapters 4, 5, 9

Moving Averages and Exponential Smoothing methods are introduced. Decomposing time series data into trends, cycles, and seasonality’s are discussed. Re-visit exponential smoothing and forecasting. Autoregressive Integrated Moving Averaging (ARIMA) models are reviewed. The Box-Jenkins method of model estimation, diagnostic checking, and forecasting are presented and evaluated. Combinations of regression and ARIMA (i.e., transfer functions) are explored. Comprehensive economic/financial models are developed using these methodologies. More advanced demonstrations of MINITAB and ECONOMAGIC are offered.

Topic 7: Data/Text Mining and Forecasting
Handout on Blackboard (Ereserves)

A brief introduction on Data and Text Mining and its applications to forecasting issues. Three techniques of Data Mining are introduced—Classification Trees, Regression Trees, and Neural Networks. Examples are drawn from bank fraud, financial forecasting, student retention rates, and stylography, etc. A brief introduction to Data and Text Mining computer software in SAS Enterprise Miner is introduced.

Topic 8: Current Issues in Forecasting

Topical issues are discussed and tailored toward specific interests of the student. The topics are selected in accordance with the research interests of the student.

SPECIFIC COURSE OUTLINE for SPRING 2017 SEMESTER

Lecture #1: Review of Statistics–(January 23-February 2)

Chapter 2 is a basic review of the elementary statistics needed for the remainder of the course. Math 117, or its equivalent, should have covered most of the material but a brief refresher on your part is time well spent. Besides the summary statistics of the mean, variance, and standard deviation, you should look at the t-test and hypothesis testing. Both are used extensively in the course.

Lecture #2: Introduction to Time Series Data–(February 2-February 9)
Chapter 3 explores data both graphically and from an analytic perspective. Most of the chapter’s focus is on ‘time series’ data; that is, data gathered over a specific period of time (i.e., 1950—2017). Forecaster’s attempt to decompose these data into several unique components such as a trend, cycle, seasonal, and random elements in order to determine the nature of the information contained in the data. In its simplest form, these decomposed data can form the underlying basis for a forecasting model that separately identifies the trend, cyclical, seasonal, and random components.
Lecture #3: Stationarity and Differencing–(February 9-Febuary 16)
A continuation of Chapter 3, this lecture’s focus deals with the issue of stationarity. Without getting into too many details, stationarity is an important aspect of time series data and impacts the ability of quantitative techniques to forecast in a reliable and accurate fashion. Strictly speaking, if the data are non-stationary, the techniques discussed here are questionable with regards to forecasting (on a more advanced level, the issue of stationarity closely corresponds to the absence/presence of a unit root). However there are statistical procedures that allow us to circumvent this stationarity issue (i.e., differencing).
Lecture #4: Simple Linear Regression Analysis–(February 16-Febuary 23)
Chapter 6 is a key chapter for the forecasting course and as such we will spend a considerable amount of time reviewing its contents. Now we shift attention to a combination of two variables—one called a dependent variate (Y), the other the independent or explanatory variable (X). Information flows from X to Y and the analyst must establish this direction of causality. Economic or financial theory can help identify which variable is dependent and which ones are independent. For example, in economics the theory of the consumption function suggests that disposable income is the independent variable (X) and personal consumption is the dependent variable (Y). The relationship is expected to be direct or positive; tell me how much you earn (X) and I’ll predict how much you will spend (Y). Another well behaved financial causal relationship between two variables is money supply (X) and interest rates (Y). According to the speculative demand for money established through liquidity preference theory, as the money supply increases, interest rates are expected to decline. This causality implies an inverse relationship between the X and Y variables.
Lecture #5: Non-Linear Regression Analysis in the Variables–(February 23-February 30)

Towards the end of Chapter 6, Hanke and Wichern begin a discussion of non-linear or curved functions. These are important not only for estimation purposes but for theoretical reasons as well. Economists rarely talk of demand lines but rather refer to demand curves, they prefer the term Phillips curve rather than a Phillips line, and so on. Economists are also likely to display results in terms of elasticity’s and certain nonlinear specifications lend themselves to this interpretation.

Lecture #6: General Review of the Simple Regression Model–(February 30-March 2)

This lecture concludes Chapter 6 on the Simple Linear or bi-variate Regression analysis. The next topic covered is Multiple Regression and I want to be sure that we have a firm understanding on the bi-variate or simple regression model before proceeding. You should be able to estimate and interpret a regression equation. By that I mean, through using MINITAB you should demonstrate competence in implementing the program and interpreting the relevant output. Interpretation includes an analysis of the regression coefficients (both slope and vertical intercept), the relevant hypothesis tests (i.e., t-tests and the F-test), and the R-squared. You should be capable of an analysis of the residuals in order to determine normality and randomness in the error term. And finally you should be able to forecast your results, both a point and an interval forecast, several time periods into the future. In addition, you should also be capable of estimating and interpreting a double logarithmic regression specification.

Lecture #7: Multiple Regression Model–(March 2-March 9)

We have now completed the simple regression model and begin with an introduction to Multiple Regression estimation in Chapter 7. The major differences between the bi-variate and multi-variate models are additional independent variables that are included in the regression model; that is, we still maintain a single dependent variable (Y) but now include several independent variates (X’s). One reason a considerable amount of time was spent on the bi-variate model is, with the exception of minor modifications in the interpretation of the regression coefficients themselves, most explanations of the remaining supporting statistics are virtually the same. That is, the t-tests and F-tests still establish statistical significance, the R2 measures the goodness of fit, and point and interval forecasts are computed exactly the same as before.

Lecture #8: Mid-term Exam—(March 20—March 27)
I will post a multi part Mid-term Exam on Monday March 20th and the due dates are Friday March 24th for Part 1 and Monday March 27th for Part 2. More details on the Mid-term will become available March 20.
You should start reviewing Chapters 1, 2, 3, 6, and 7. Also to assist in your preparation, revisit my lecture notes and videos.
Lecture #9: Extensions of the Regression Model—Dummy or Dichotomous Variables (March 27—April 4)

Chapter 7 also introduces the idea of incorporating qualitative information into a regression model by introducing Dummy or Dichotomous variables. For example, let’s assume when developing a Human Capital model of wage determination, in addition to the usually suspects for independent variables such as years of schooling, on-the-job training, etc., you would want to estimate the impact gender has on wages. Dummy or dichotomous variables allow for quantitative estimates of these qualitative characteristics.

Lecture #10: Extensions of the Regression Model—Multicollinearity (April 4—April 11)

Chapter 7 introduces the idea of multicollinearity in a regression model. Our approach is to analyze the causes, consequences, test-to-detect, and correction procedures this issue brings to our regression models. This issue only occurs within the independent variable data set and as such is related to multiple regression modelling; that is, what happens to our regression estimates if the independent variable themselves are linearly related to each other?
Lecture #11: Extensions of the Regression Model—Autocorrelation (April 11—April 18)

Chapter 8 introduces the statistical concept of autocorrelation in a regression model. Our approach is to again analyze the causes, consequences, test-to-detect, and correction procedures this issue bring to regression models. Where multicollinearity is an issue with the independent variables, autocorrelation takes a close look at the residuals or errors estimated from the regression model. The concern here is to determine if patterns exist in the residuals and if so how does this impact the regression estimates.

Lecture #12: Time Series Models– (April 18—April 25)

Chapter 8 proposes the use of TIME as an independent variable. Let’s say you’re required to forecast company sales and it’s needed within the next hour. A fully multiple regression model with several independent variables may not be feasible. One option is to use TIME as an independent variate to at least estimate the trend component in company sales. This lecture will illustrate how trends are estimated and will also demonstrate how, through the use of dummy variables, a company’s seasonal indexes can be calculated.

Lecture #13 Final Paper—(April 25 – May 9)

A description Your Research Paper and how the paper should be organized will be distributed in April. Some past students have included this paper as an exhibit in their e-portfolio. A quality effort could be helpful when applying for a job since, if done properly, this paper is an excellent showcase of your accomplishments here at Pace University. Employers, especially those requiring quantitative skills, are quite impressed by the quality of the work shown in these papers.

Quality of Homework Assignments

You will receive a grade for each of the Homework assignments and what follows is a brief description of exactly how you are graded. In either case you must submit your assignments on time in order to receive credit.

A or A- B+ or B B- or C+ C or C- D+ / D / D-
Timeliness Responses are always posted on time. Responses are on time – one occasionally missed. Responses usually on time or occasionally missed Responses missed or late more than once. Responses consistently posted late or missed.
Questions Provides provocative questions on time that help us rethink readings. Questions adequate but not provocative. “Surface” questions that provide not critical analysis Questions off base, inappropriate, or irrelevant for readings. Doesn’t post questions on time.
Thorough Consistently addresses all parts of questions well. Usually all parts addressed, some better than others. Sometimes some parts are missing but what is posted is usually thorough. Responses to questions are often quite incomplete. Responses consistently address questions minimally.
Thoughtful Connections Consistently responses make
thoughtful & specific connections to readings. Links theory to practice and across posts, readings, or discussions. Usually thoughtful connections to readings but often lacking
connections made across posts, readings, or class discussions. Sometimes responses do not demonstrate an under-standing of readings.
Connections may be forced or incomplete. Often responses are lacking in substance and, quite possibly,
inaccurate. Connections made may be inaccurate or incoherent. Responses are consistently done in a hasty fashion with no thoughtful connections
made.
Substantive Responses to Classmates Often reads others’ posts and offers informed questions, comments, & connections. Usually comments on others’ posts. Interesting comments but they may not indicate clearly a close read. Sometimes comments on others’ posts. Comments are sometimes off base – not clearly relating to others’ posts. Infrequently comments on others’ posts and comments are sometimes inappropriate – unhelpful, not constructive, rude. Rarely comments on others’ posts or, when does, comments are often inappropriate – unhelpful, not constructive, rude.

Other Important Information.

Students must accept the responsibility to be honest and to respect ethical standards in meeting their academic assignments and requirements. Integrity in the academic life requires that students demonstrate intellectual and academic achievement independent of all assistance except that authorized by the instructor. The use of an outside source in any paper, report or submission for academic credit without the appropriate acknowledgment is plagiarism. It is unethical to present as one’s own work, the ideas, words or representations of another without the proper indication of the source. Therefore, it is the student’s responsibility to give credit for any quotation, idea or data borrowed from an outside source.

Students who fail to meet the responsibility for academic integrity subject themselves to sanctions ranging from a reduction in grade or failure in the assignment or course in which the offense occurred to suspension or dismissal from the University. Students penalized for failing to maintain academic integrity who wish to appeal such action may petition the department chair to request a hearing on the matter.

Pace University believes that it is important that students receive appropriate accommodation for any disability. If you have a disability for which you are or may be requesting an academic accommodation, you must register with the Coordinator of Services for Students with Disabilities. Trained professional Counselors will:

-Evaluate your medical documentation;
-Conduct appropriate tests or refer you for same;
-Make recommendations for your plan of accommodation; and
-Contact your professors (with your permission) to arrange for the
recommended accommodations.

Your professor is not authorized to provide any accommodation prior to you arranging for same through the Counseling/Personal Development Center. If you have, or believe you have, a disability, be sure to follow the above procedure.