Short description of project
This project will investigate and compare some forecasting methods for modeling time series in economics and finance. These methods include time series regression (with heteroscedasticity and autocorrelation correction, HAC), autoregressive integrated mobbing average (ARIMA), and neural networks (NN) models. This research will study the effectiveness of the forecasting performance, particularly to answer whether a complex method always gives a better forecast than a simpler method. The research continues with some hybrid models, by the inclusion of intervention effect or the use of preprocessing data, in order to improve forecasting accuracy.
The following time series will be studied:
1. US Enplanement (international and/or domestic) – to evaluate the forecasting performance of the applied models by using different preprocessing techniques.
3. Hong Kong and Macau Tourists – tom formulate the forecasting models for Hong Kong and Macau tourist’s arrivals and investigate whether intervention analysis can improve the forecast accuracy.
4. BRIC (Brazil, Russia, India, & China) Stock Market Indexes – to examine if nonlinear neutral networks models can outperform the two well- known financial forecasting methods, capital asset pricing model (CAPM) and Fama and French’s (FF’s) 3-factor model in these four emerging markets.
The following time series will be studied:
1. US Enplanement (international and/or domestic) – to evaluate the forecasting performance of the applied models by using different preprocessing techniques.
3. Hong Kong and Macau Tourists – tom formulate the forecasting models for Hong Kong and Macau tourist’s arrivals and investigate whether intervention analysis can improve the forecast accuracy.
4. BRIC (Brazil, Russia, India, & China) Stock Market Indexes – to examine if nonlinear neutral networks models can outperform the two well- known financial forecasting methods, capital asset pricing model (CAPM) and Fama and French’s (FF’s) 3-factor model in these four emerging markets.
Information of Offered Internship
Level of Internship Hours per Month
Level 3 - 60 hours
Commencement Month
June
Duration
6 Months
Internship requirements: i.e. work, practice and training
- Good logical & analytical mind.
- Accounting/Finance students are preferred.
On the job-training will be provided to successful candidate, including:
- Operation of two databases: Datastream (global market) and CS
- Accounting/Finance students are preferred.
On the job-training will be provided to successful candidate, including:
- Operation of two databases: Datastream (global market) and CS