Econ6037: Economic Forecasting
Spring 2014, University of Hong Kong
Assignment #3 Regression Based Forecast
Due date: Wednesday, April 23, 11:30p.m. (via the course website)
A note from the instructor
1. This assignment is meant to be group work. The grou
library(sandwich)
library(lmtest)
# - The estimation requires a package named "vars".
library(vars)
rm(list = ls()
# remove (almost) everything in the working environment.
# You will get no warning, so don't do this unless you are really sure.
#rm(lis
ECON 6037
Project 2-US
Te Liu
Forecasting Unemployment Rate of US
LIU, Te
2013963598
26/03/2014
0
Introduction
Unemployment rate is a very common index to measure the health of the economy, it is
very important for not only companies, but also for individ
Econ6037: Economic Forecasting
Spring 2014, University of Hong Kong
Assignment #2 Forecasting Unemployment rate (monthly or quarterly)
Due date: Wednesday, March 26, 11:30p.m. (via the course website)
A note from the instructor
1. This assignment is meant
#GDPforecast
data=read.csv(file="/Users/Yvette/Documents/Master/Economic
Forecasting/PS01/final data.csv",head=TRUE,sep=",") # read in the data in CSV
format
names(data)#displaythenamesofthevariablescontainedinthedataframe
summary(data)#producesummarystat
Evaluating and Combining Forecasts
Ka-fu WONG
University of Hong Kong
April 15, 2014
kf011
Evaluating and Combining Forecasts
Contents
1. The Evaluation Situation
2. Forecast Schemes
2.1. Fixed scheme
2.2. Recursive scheme
2.3. Rolling scheme
2.4. Fixed,
An example of regression-based forecast
(CHEN, 2013)
Ka-fu WONG
University of Hong Kong
April 10, 2014
Ka-fu WONG
Chen (2013): Forecasting Oil Prices
Contents
1.
2.
3.
4.
5.
6.
7.
8.
Paper
Purpose
Model for forecasting nominal crude oil prices
Model for f
Forecasting Turning Points
Ka-fu WONG
University of Hong Kong
May 8, 2014
KF015
Forecasting Turning Points
Contents
1.
2.
3.
4.
5.
Recession Variable
Linear Probability Model
Logit and Probit
Estrella and Mishkin (1998)
Additional readings
4
5
6
9
21
/ 2
Unit Roots, Stochastic Trends, ARIMA Forecasting Models
Ka-fu WONG
University of Hong Kong
April 3, 2014
kf008
Forecasting with Regression Models
Contents
1. A quick review of autoregressive models
2. Random walk as a unit root
3. Drastic dierence between
Instability and Structural Breaks
Ka-fu WONG
University of Hong Kong
May 8, 2014
KF014
Structural Breaks
Contents
1. Structural Break
1.1. A scenario
1.2. Problem How to estimate 0,2 and 1,2 ?
2. Testing for Structural Change of Known Timing
3. Testing fo
# To simulate a set of VAR data and hence
# to understand how to estimate VAR models with the "vars" procedures.
# - The estimation requires a package named "vars".
library(vars)
# remove (almost) everything in the working environment.
rm(list = ls()
Econ6037: Economic Forecasting
Spring 2014, University of Hong Kong
Assignment #4 Forecasting Exchange Rate
Due date: Wednesday, May 14, 11:30p.m. (via the course website)
A note from the instructor
1. This assignment is meant to be group work. The group
Econ6037: Economic Forecasting
Spring 2014, University of Hong Kong
Assignment #1 Forecasting GDP (annual)
Due date: Wednesday, February 26, 11:30p.m. (via the course website)
A note from the instructor
1. This assignment is meant to be a group work. The
Volatility Measurement, Modeling, and Forecasting
Ka-fu WONG
University of Hong Kong
April 24, 2014
KF012
Modeling Volatility
Contents
1. Importance of volatility
1.1. Asset allocation
1.2. Carry trade
1.3. Impact of Volatility on Macroeconomy
2. Clusteri
Smoothing for Forecasting
Ka-fu WONG
University of Hong Kong
April 10, 2014
kf010
Smoothing for Forecastings
Contents
1. Use of Smoothing to produce forecast
2. Idea
3. Moving average smoothing
3.1. Two-sided moving average:
3.2. One-sided moving average:
In-class exercise Property Market in Hong Kong
Name:
Student ID No.:
(Firstname)
(Lastname)
Imagine we are writing a forecast report about the Hong Kongs property market. The following is a plot of
the price indices of private residential properties. (Bas
CHAPTER 8
Pulling things together
In previous chapters, we have learned how to forecast trend component, seasonality component and
cyclical component separately. The ultimate goal is to model a general time series that consists of all
three components. Tr
In-class exercise Expectations and Covariances
Name:
Student ID No.:
(Firstname)
(Lastname)
Consider the random variables X and Y with the following expectations, variances and covariances.
E(X) = 3, V ar(X) = 2
E(Y ) = 5, V ar(Y ) = 1
Cov(X, Y ) = 2
C