My Dissertation Proposal
1 like893 views
This dissertation proposal explores the properties of Philippine remittances, particularly their cyclical nature and macroeconomic impacts. It involves detailed data treatment and analysis techniques to understand the dynamics of remittances from overseas Filipino workers. The research aims to apply new methodologies to remittance data, addressing issues like exchange rate effects and structural breaks in the data.
My Dissertation Proposal
- 1. Properties of Philippine Remittances (An Econometric Analysis) Dissertation Proposal
- 2. This Presentation will touch upon: The Motivation for the research. A description of the data Treatment. Initial departures and preliminary Results. Issues and Directions to explore.
- 3. The Motivation for the research.
- 4. Remittances are particularly significant for the Philippines and for Filipinos .
- 5. The Filipino Diaspora and Workers’ Remittances 15 10 5 0 2000 2001 2002 2003 2004 2005 2006 2007 Overseas Filipinos Workers’ Remittances (in Millions of Filipinos) (in Nominal US$ Billions) Sources: International Labor Organisation (2009) and Bangko Sentral ng Pilipinas (2008)
- 6. The Seven Largest Remittance Recipient Countries India China Mexico Philippines France Spain Germany 0 15 30 45 60 Remittances % of GDP (in Nominal US$ Billions) (2008 GDP) Source: The World Bank Migration and Remittances Data (November 2009)
- 7. “Stylized facts”about remittances SOME tions contrast with findings from individual country cases. Ob serva e abo ut th e lite ratur S) Some approaches have yet to be applied TANCE (ON REMIT to the Philippine case. Many studies have approached the topic from a development perspective. Of the approximately 102,000 articles on remittances available on Google Scholar, not a single one was written by me!
- 8. Apply new approaches to Philippine Data. Inquire into the cyclicality and other properties of remittances to the Philippines. Approach the topic from a - macroeconomic perspective. Develop the research like an Wh open-source project. y No t.. .?
- 9. The Treatment applied to the data
- 10. Data for Overseas Filipino Workers’ (OFW) remittances and other macroeconomic variables were obtained for 1989Q1 to 2008Q4. Each time series (v) was assumed to have the following components: vt = trend t + cyclical t + seasonal t + et Three techniques were tried to remove the seasonal component of the data. Two business cycle filters were considered to remove the time trend and isolate the cyclical component of the data.
- 11. ln(y) (Raw Data) ln(r) (Raw Data) 5.0 7.4 4.5 7.2 4.0 7.0 y r 3.5 6.8 3.0 6.6 1990 1995 2000 2005 1990 1995 2000 2005 Quarterly Frequency, 1989Q1-2004Q4 1989Q1-2008Q4 Quarterly Frequency, 1989Q1-2004Q4 1989Q1-2008Q4
- 12. HP Filtered y by Seasonal Adjustment Technique CF Filtered y by Seasonal Adjustment Technique 0.04 0.02 0.02 Growth Rate (Percent) Growth Rate (Percent) 0.00 0.00 -0.02 -0.02 -0.04 -0.04 by LOESS by LOESS by Moving Averages by Moving Averages by Dummy Variables by Dummy Variables 1990 1995 2000 2005 1990 1995 2000 2005 Quarterly Frequency: 1989Q1 to 2008Q4 Quarterly Frequency: 1989Q1 to 2008Q4 Percentages are Normalized to 1 Percentages are Normalized to 1
- 13. HP-Filtered r by Seasonal Adjustment Technique CF Filtered r by Seasonal Adjustment Technique 0.4 0.6 0.3 0.4 0.2 Growth Rate (Percent) Growth Rate (Percent) 0.2 0.1 0.0 0.0 -0.1 -0.2 -0.2 by LOESS by LOESS -0.4 by Moving Averages by Moving Averages by Dummy Variables by Dummy Variables -0.3 1990 1995 2000 2005 1990 1995 2000 2005 Quarterly Frequency: 1989Q1 to 2008Q4 Quarterly Frequency: 1989Q1 to 2008Q4 Percentages are Normalized to 1 Percentages are Normalized to 1
- 14. Preliminary Results thus far from work on the topic
- 15. First, the research explores the cyclicality of workers’ remittances to the Philippines. In keeping with the literature, the idea was to develop regression models using variables relevant to remittance sending: r = f( GDP , Past remittance , Exchange Rate )
- 22. Conjecture: Perhaps the different remittance-sending behavior described by the models is due to the type of OFW deployed in each particular country.
- 24. Vector autoregressions were employed with the end in view of understanding remittances’ macroeconomic impact. A lag length of one was used to prevent overfitting using VARs. Impulse-response functions were also employed to model the effect of shocks to the variables.
- 25. IRFS USING HP-FILTERED DATA Orthogonal Impulse Response from hp.ya Orthogonal Impulse Response from hp.ra 0.15 0.02 0.10 hp.ya hp.ya 0.00 0.05 -0.02 0.15 0.00 xy$x xy$x 0.02 0.10 hp.ra hp.ra 0.00 0.05 -0.02 0.00 0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10
- 26. IRFS USING CF-FILTERED DATA Orthogonal Impulse Response from cf.ya Orthogonal Impulse Response from cf.ra 0.06 0.02 0.04 0.01 cf.ya cf.ya 0.02 0.00 0.00 0.06 -0.02 0.02 xy$x xy$x 0.04 0.01 cf.ra cf.ra 0.02 0.00 0.00 -0.02 0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10
- 28. Issuesand Directions to explore
- 29. Why are there different results between different filtering techniques? Can the exchange rate effects be accounted for in the analysis? How to handle structural breaks in the data? How do remittances measure up against other inflows like FDI? What other macroeconomic variables would be relevant to the analysis?
- 30. END OF PRESENTATION T H A N K Y O U Brian L. Belen bbelen@gmail.com brianbelen.blogspot.com @brianbelen