Options
Harrison, Michael J.
Preferred name
Harrison, Michael J.
Official Name
Harrison, Michael J.
Research Output
Now showing 1 - 6 of 6
- PublicationExploring nonlinearity with random field regression(University College Dublin. School of Economics, 2007-11)
; ; Random field regression models provide an extremely flexible way to investigate nonlinearity in economic data. This paper introduces a new approach to interpreting such models, which may allow for improved inference about the possible parametric specification of nonlinearity.488 - PublicationNonlinearity as an explanation of the forward exchange rate anomaly(University College Dublin. School of Economics, 2008-01)
; ; ; This paper shows that nonlinearity can provide an explanation for the forward exchange rate anomaly (Fama, 1984). Using sterling-Canadian dollar data, and modelling nonlinearity of unspecified form by means of a random field, we find strong evidence of time-wise nonlinearity and, significantly, obtain parameter estimates that conform with theory to a high degree of precision: the anomaly disappears.331 - PublicationA flow analysis of the link between Irish and British unemploymentThis paper is a contribution to the research on Irish unemployment which for the first time models the flows into and out of the Live Register. Using the quarterly flow data contructed by the authors (see A Flow Analysis of the Irish Live Register,Economic and Social Review, Volume 26, pp. 45-58, 1994), the analysis proceeds within a small open labour market framework, making use of the concepts of cointegrations and error-correction to model the flows and hence the migratory movements between Ireland and Britain. We outline the advantages of using flow data to link unemployment in a small region and a large region within an integrated labour market. We show that demographic changes resulting from natural increases in population and migration anre likely to be the key determinants of unemployment turnover in Ireland. We conclude that any explanation of Irish unemployment must account for these special features of the economy, and in particular must indicate why domestic employment movements seem to have had so little effect on the unemployment flows.
169 - PublicationModelling Ireland’s exchange rates : from EMS to EMU(University College Dublin. School of Economics, 2007-11-17)
; ; This paper attempts to model the nominal and real exchange rate for Ireland, relative to Germany and the UK from 1975 to 2003. It offers an overview of the theory of purchasing power parity (Ppp), focusing particularly on likely sources of nonlinearity. Potential difficulties in placing the analysis in the standard I(1)/I(0) framework are highlighted and comparisons with previous Irish studies are made. Tests for fractional integration and nonlinearity, including random field regressions, are discussed and applied. The results obtained highlight the likely inadequacies of the standard cointegration and Star approaches to modelling, and point instead to multiple structural changes models. Using this approach, both bilateral nominal exchange rates are effectively modelled, and in the case of Ireland and Germany, Ppp is found to be valid not only in the long run, but also in the medium term.197 - PublicationExploring long memory and nonlinearity in Irish real exchange Rates using tests based on semiparametric estimation(University College Dublin. School of Economics, 2009-01-15)
; ; Deciding whether a time series that appears nonstationary is in fact fractionally integrated or subject to structural change is a diffcult task. However, various tests have recently been introduced for distinguishing long memory from level shifts and nonlinearity. In this paper, three testing approaches based on the properties of semiparametric estimators of the fractional differencing parameter, d, are described and applied to the (log) Ireland-United Kingdom and Ireland-Germany real exchange rates. The two exchange rates behave quite differently over time and the new tests give different results for each; but overall the results provide fairly strong support for the possibility of nonlinearity rather than long memory.108 - PublicationThe impact of discriminatory legislation on Irish female unemployment flows(Trinity College Dublin. Department of Economics, 1995)
; ; Ireland provides us with a unique case-study of the effects of discrimination in the labour market. Since the ninteen-sixties and until the late nineteen-eighties, gradual reforms of explicit discrimination against females with regard to entitlement to and duration of unemployment assistance and benefit have been introduced. The primary aim of this paper is to asses the impact that these reforms have had on the level of female turnover activity in the Live Register. The results show that the reforms may be modelled as well defined discrete shifts in the inflows and it is noteworthy that the more significant of the estimated effects of reforms are those corresponding to those which gave the large numbers of females that were in non-activity the option of entering the Live Register without any prior need of employment contributions. The results also provide evidence of a secondary effect of reforms on the level of female outflows, and appear to support the hypothesis that the reforms have encouraged females to remain on the Live Register for longer periods of time.106