I was thinking of using a VAR model. can you direct me to other similar studies
Hi! Depends on the time series behaviour (linear or non-linear patterns). In the case of linear behaviour, such as ARDL or bivariate VAR/VECM would be ok, but if non-linear pattern has been detected (with linearity tests), something like SETAR, TVAR/TVECM, GAM would also be ok..... Moreover, modelling (especially, under non-linear patterns) with advanced machine learning techniques would be applicable...
It seems that many econometricians foreget the first part of "econometrics", i.e. they foreget economics. VAR, ARDL, ECM etc. are not economic models, but statistic methods to test economic hypotheses (theories, models). Before applying them, one should put much effort in the (principal) specification of an economic equation/model. The decision what statistical method to apply will depend on this hypothesis and the available data. One should try to start with the simplest methods (e.g. comparing the development of variables).
By the way, these methods can be and are applied in other sciences. For example, Gauss applied OLS in 1801 to find-out the orbit of Ceres, which was discovered some years before. But without Kepler's and Newton's theories (laws) he could not do that. These laws gave him the information about what function to estimate (i.e. specification).
It depends on the nature of the time series data. If there is existed structural breaks than non linear analysis is favourable. Usually time series data has many fluctuations so you may use NARDL econometric approch.
Can a time-series predict economic activity?Can a time-series predict economic activity?Time series seems to describe the law that cannot change. Once the law changes, the development trend of things will have no inertia, and the time series cannot predict this unexpected change.
Muhammad Asif Amjad
Structural mutations, the nonlinearity is also useless? For example, the "black swan" has appeared.
I always think that it is not enough for economics to describe laws only with mathematical models, because economic laws will change, and mathematical models can only describe fixed laws, including differential equations and probability models. So I proposed another way to describe the laws of economics: program.
Preprint Analysis of Financial Crisis Causes and Complex Systems Scie...
There are several time series models that can be used to measure the impact of unemployment and labour market dynamics on poverty. One such model is Vector Autoregression (VAR). VAR models are used to analyze the dynamic relationship between multiple time series variables. Another model is Autoregressive Integrated Moving Average (ARIMA). ARIMA models are used to analyze the time series data and forecast future values.
- Badges
- Science topic