ECON 112 UCSD Unemployment and Inflation Rate in Belgium Data Analysis

Description

2 attachmentsSlide 1 of 2attachment_1attachment_1attachment_2attachment_2.slider-slide > img { width: 100%; display: block; }
.slider-slide > img:focus { margin: auto; }

Unformatted Attachment Preview

Econ 112: Macroeconomic Data Analysis
Writing Assignment 1
To write a concise report explaining why and how you applied the Box-Jenkins methodology to two selected
data sets. Explain what choices you made, and why, in applying the model to the data sets.
Objectives
This assignment has two objectives. The first objective is to have you practice how to decide what is essential
and what it is not essential when you have to choose what numbers and figures to report out of your statistical
analysis. The second objective is to practice how to best express in writing the information that you recognize
in the numerical results and figures that you choose to report.
Text of Assignment
Suppose you are a junior research analysis at the International Monetary Fund, and you are part of a team of
economists that study the country you have been assigned in class. You receive the following email from your
Dear your name, at our next team meeting we will be discussing the effects of fiscal and monetary policies
on the unemployment rate and the inflation rate of our assigned country. It would be useful for us to have
a written report on the statistical dynamic models that best describe those two data series. Because of your
expertise with time series, I would like to ask you to prepare the report by February 1. Thank you.
Your report should include of all the following elements.
1. Introduction.
State what you aim to achieve in the report and summarize your findings.
2. Data.
Outline what data you are using, present summary statistics and plot data that you think are important.
3. Preferred Specification.
Summarize the Box-Jenkins methodology steps outlined in class. Articulate how you identified your preferred
specifications for the Unemployment Rate and Inflation. Only results from your preferred specifications should
be included. Use well-organized tables to report estimates and figures to show important patterns in your
data (ACF, PACF). Edit tables and figures so that they are easy to read (no raw R output should ever be
reported).
4. Limitations.
Describe the limitations of your preferred specification. Tell the reader what the potential weaknesses are for
1
analysis (such as sample length, information criteria, etc.)? In light of the robustness considerations, are there
alternative specifications that should be considered as close substitutes to the one you proposed?
5. Future Work.
Suggest possible directions for further analysis.
Note: The body of the report should be at least 1,250 words, excluding tables and figures.
2
Box-Jenkins Procedure to Estimate ARMA Models
1. Identifification Stage
 analyze plot of raw data
 analyze ACF and PACF
 identify plausible models
2. Estimation Stage
 estimate plausible models
 evaluate AIC, BIC
 take a look at p-value of coefficients
 select a parsimonious model that has good fit
3. Diagnostic Checking
 plot residuals of chosen model
 look for periods of poor fit
 analyze standardized residuals, ?t/??, they should look Gaussian