Simple regressions with linear time trends
The author considers the following problem. Let \(x_t\) be a scalar trend stationary time series \(x_t =\gamma_x+\mu_x t+x_t^0\), \(\mu_x \neq 0\), \(t=1, 2,\ldots,T.\) The purely stochastic deviations from the linear trend are weakly stationary with zero mean. The scalar series \(y_t\) is given by \(y_t = a+b x_t +u_t\), \(t=1, 2,\ldots,T\), where the error \(u_t\) is again a zero-mean weakly stationary process that may be correlated with \(x_t^0.\) The problem is to find the limiting distribution of the ordinary least squares (OLS) estimator of \(a\) and \(b\). Let NEWLINE\[NEWLINE\hat b-b=\sum(x_t-\bar x)u_t/\sum (x_t-\bar x)^2,NEWLINE\]NEWLINE NEWLINE\[NEWLINE\omega_u^2=V(u_t)+2\sum_{\tau=1}^{\infty} \text{cov} (u_t, u_{t-\tau}),\quad T^{-1}\sum (u_t -\bar u)^2 \Rightarrow V(u_t),NEWLINE\]NEWLINE NEWLINE\[NEWLINE\hat u_t=y_t -\bar y - \hat b(x_t -\bar x),\quad s^2 =T^{-1}\sum \hat u_t^2,NEWLINE\]NEWLINE NEWLINE\[NEWLINER^2=1-s^2/T^{-1} \sum (y_t - \bar y)^2,\quad \hat\rho_u =(T^{-1}\sum_{t=2}^T \hat u_t \hat u_{t-1})/s^2,NEWLINE\]NEWLINE NEWLINE\[NEWLINEt_b=(\hat b -b) \bigl\{\sum (x_t - \bar x)^2 \bigr\}^{1/2}/s.NEWLINE\]NEWLINE The main result of this paper is the following. Under some assumptions on the OLS regression \(y_t = \hat b x_t + \hat u_t\), \(t=1, 2, \ldots, T\), the following assertion is true: NEWLINE\[NEWLINET^{1.5}(\hat b -b) \Rightarrow N\biggl(0, 12 \omega_u^2/\mu_x^2\biggr),\quad t_b \Rightarrow N\biggl(0, \omega_u^2/\sigma_u^2\biggr),NEWLINE\]NEWLINE NEWLINE\[NEWLINEs^2 {\buildrel P \over \longrightarrow} \sigma_u^2,\quad \hat{\rho_u} {\buildrel P \over \longrightarrow} \text{cov} (u_t, u_{t-1})/\sigma_u^2, \quad T^2 (1-R^2) {\buildrel P \over \longrightarrow} 12 \sigma_u^2/\mu_x^2 b^2,NEWLINE\]NEWLINE as \(T\to \infty\) with \(\sigma_u^2 = V(u_t)\).NEWLINENEWLINENEWLINEThe author concludes that when a linear trend dominates the stochastic components the rates of convergence and the limit distributions of OLS statistics are exactly the same as in the case of cointegrated regressions with drift. In particular, the asymptotic standard normal \(t\) statistics are available. The asymptotic inference requires no distinction between simple regressions of trend stationary series of cointegrated variables with drifts.
- Fractional cointegration in the presence of linear trends
- Regression with integrated regressors
- Linear regression with deterministic regressors and unit root in the variance
- OLS-BASED ASYMPTOTIC INFERENCE IN LINEAR REGRESSION MODELS WITH TRENDING REGRESSORS AND AR(p)-DISTURBANCES
- On the consequences of trend for simultaneous equation estimation
- A note on the distribution of the least squares estimator of a random walk with a linear trend
- Regression with integrated regressors
- Spurious regressions with stationary processes around linear trends
- A simple solution for spurious regressions
- Regression with slowly varying regressors and nonlinear trends
- Linear regression with deterministic regressors and unit root in the variance
- Spurious instrumental variables
- scientific article; zbMATH DE number 3907632 (Why is no real title available?)
- On the consequences of trend for simultaneous equation estimation
- Ordinary least squares estimation of simultaneous equation systems with trended data: further results
- Simple diagnostic tools for inverstigating linear trends in time series
- OLS-BASED ASYMPTOTIC INFERENCE IN LINEAR REGRESSION MODELS WITH TRENDING REGRESSORS AND AR(p)-DISTURBANCES
- scientific article; zbMATH DE number 6843406 (Why is no real title available?)
- Simple linear regression with multiple level shifts
- Spurious regression
- Spurious regression and lurking variables
This page was built for publication: Simple regressions with linear time trends
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2742771)
