2134/17058
George Saridakis
George
Saridakis
Grammatoula Papaioannou
Grammatoula
Papaioannou
Forecasting self employment in the UK
Loughborough University
2015
Self-employment
ARMA
State space
Var
Business and Management not elsewhere classified
2015-03-19 14:05:52
Journal contribution
https://repository.lboro.ac.uk/articles/journal_contribution/Forecasting_self_employment_in_the_UK/9501623
In this paper we forecast UK self-employment, using annual data for five decades. We use the autoregressive moving average (ARMA) methodology to produce a forecast three periods into the future (2014-2016). We also express the ARMA model as a state space model and estimate one-step predictions and dynamic forecasts for the same period. We then compare the univariate forecasts with multivariate multi-step ahead forecasts using a vector autoregressive (VAR) methodology. Comparing the multivariate forecasts with the univariate forecasts, we observe that both point to an increase in UK enterprise activity in the future, with the increase being sharper in the former.